Non-enzymatic and highly sensitive lactose detection utilizing graphene field-effect transistors

https://doi.org/10.1016/j.bios.2020.112419Get rights and content

Highlights

  • A graphene field-effect transistor (G-FET) biosensor was demonstrated for highly sensitive and specific lactose detection.

  • Graphene was functionalized with an engineered mutant of the carbohydrate recognition domain (CRD) of human galectin-3.

  • The sensor had a limit of detection (LoD) of 200 aM and was specific to lactose over a dynamic range of 1 fM to 1 pM.

Abstract

Field-effect transistor (FET) biosensors based on low-dimensional materials are capable of highly sensitive and specific label-free detection of various analytes. In this work, a FET biosensor based on graphene decorated with gold nanoparticles (Au NPs) was fabricated for lactose detection in a liquid-gate measurement configuration. This graphene device is functionalized with a carbohydrate recognition domain (CRD) of the human galectin-3 (hGal-3) protein to detect the presence of lactose from the donor effect of lectin – glycan affinity binding on the graphene. Although the detection of lactose is important because of its ubiquitous presence in food and for disease related applications (lactose intolerance condition), in this work we exploit the lectin/carbohydrate interaction to develop a device that in principle could specifically detect very low concentrations of any carbohydrate. The biosensor achieved an effective response to lactose concentrations over a dynamic range from 1 fM to 1 pM (10−15 to 10−12 mol L−1) with a detection limit of 200 aM, a significant enhancement over previous electrochemical graphene devices. The FET sensor response is also specific to lactose at aM concentrations, indicating the potential of a combined lectin and graphene FET (G-FET) sensor to detect carbohydrates at high sensitivity and specificity for disease diagnosis.

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)

  • A.A. Cagang et al.

    Anal. Chim. Acta

    (2016)
  • B. Cai et al.

    Biosens. Bioelectron.

    (2015)
  • S. Chen et al.

    Biosens. Bioelectron.

    (2020)
  • T.-Y. Chen et al.

    Biosens. Bioelectron.

    (2013)
  • C.I.L. Justino et al.

    Trac. Trends Anal. Chem.

    (2017)
  • L.L. Kiessling et al.

    Chem. Biol.

    (1996)
  • Y.M. Lei et al.

    Biosens. Bioelectron.

    (2017)
  • B.H. Nguyen et al.

    Curr. Appl. Phys.

    (2016)
  • S. Viswanathan et al.

    Mater. Today

    (2015)
  • S. Zhao et al.

    J. Electroanal. Chem.

    (2019)
  • R. Campos et al.

    ACS Sens.

    (2019)
  • P.M. Collins et al.

    Acta Crystallogr. D

    (2007)
  • C. Diehl et al.

    J. Am. Chem. Soc.

    (2010)
  • X. Dong et al.

    Adv. Mater.

    (2010)
  • K. Drickamer

    Nat. Struct. Biol.

    (1995)
  • W. Fu et al.

    Adv. Mater.

    (2017)
  • Cited by (21)

    • Recent advances in nanomaterials based biosensors

      2023, TrAC - Trends in Analytical Chemistry
    • Electrochemical biosensors for biomolecules

      2023, Handbook of Biomolecules: Fundamentals, Properties and Applications
    • A 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 Confederation
      Citation 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-2
    • Systematic review on lectin-based electrochemical biosensors for clinically relevant carbohydrates and glycoconjugates

      2021, Colloids and Surfaces B: Biointerfaces
      Citation 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.

    View all citing articles on Scopus
    View full text