Take a look at the Recent articles

Abstract

We demonstrate a graphene-based sensor for lactate detection with a sensitivity of 100pM (10^-10M) and range of detection up to 200 mM (0.2 M), that is, 12 orders of magnitude, with high accuracy and rapid response times. This development holds significant promise for advancing point-of-care diagnostics and continuous monitoring in healthcare settings, facilitating early intervention and personalized treatment strategies.

Introduction

Lactate is an intermediate in cellular bioenergetics and a by-product of anaerobic metabolism [1]. Due to its close correlation with the availability of oxygen in tissues, the presence of lactate can serve as an indicator of hypoperfusion. Therefore, quantifying lactate levels may indicate patients' oxygenation state and the presence of hypoxia caused by circulatory impairments [2]. This is especially important for patients requiring critical care monitoring due to cardiomyogenic shock (arrhythmias or ventricular failure), sepsis, or acute respiratory distress syndrome [1-4]. Monitoring lactate buildup and accumulation during physical activity can minimize muscle soreness, pain, and fatigue [5]. This can support tailoring training loads and schedules to maintain athletes' blood lactate concentration within the aerobic level [2]. In addition, it can help prevent injuries, not only to improve the performance of athletes but also to improve the overall well-being of the elderly, who are engaging in a growing trend of heightened activity.

Traditionally, invasive blood tests are employed to quantify the level of lactate. Clinically, a blood sample is obtained by inserting a needle into a vein in the arm. After that, the lactate is quantified using enzymatic analysis [2]. Despite being used for decades, this technique does not capture rapid fluctuations in lactate levels and requires specialized manpower and equipment. Alternatively, hand-held blood lactate analyzers can quantify the biomarker from a blood sample obtained from the tip of the fingers [6]. Despite their popularity, these devices often have limited measurement ranges, potentially excluding lactate amounts that do not fall within the range of the equipment [7]. Hence, developing noninvasive methods for quantifying lactate, especially across a broader range of concentrations than current techniques allow, is of great interest.

Human sweat is vital for thermoregulation and can be a source for noninvasive lactate quantification. The lactate content of sweat varies in response to exertion present in the body, exceeding 100mM in exhaustive cases [8]. Invasive sensors require the insertion of a probe or device into the body, which can be uncomfortable and potentially hinder applications in which movement can lead to the dislodgement of the sensor, such as high-performance/impact sports. Therefore, noninvasive bio-sensors are becoming increasingly important for monitoring the levels of biomarkers in medical and sports applications [9,10].

Different strategies have been tried to monitor sweat lactate, with designs succeeding in quantifying lactate at different ranges, such as 1–20, 10–100, and 5–30 mM, with sensitivities ranging from 0.6 μA/mM to 2.4 μA/mM [11-14]. Nonetheless, limitations such as sensor saturation can challenge sensor reusability [15]. Significant strides have been achieved in wearable and non-contact devices [16,17]. However, continuous monitoring of biomarkers in wearables requires the consideration of several designs requirements such as choice of materials for wear ability, sensor stretchability, ability to resist motion, material biocompatibility, adhesiveness to the skin, and effectiveness of the microfluidic system to direct the collect sweat from the body to the measurement core [18,19]. Therefore, developing alternative strategies for cost-effective, noninvasive quantification of lactate is important.

We have developed a reusable single layer and three-electrode graphene sensor, with ranges covering sweat and actually exceeding it. The single layer graphene sensor can detect lactate from picomolar to several hundred micromolar concentrations; the three-electrode sensors covers a range of micromolar to hundreds of millimolar concentrations, providing a wide range unparalleled by currently available technologies. The benchtop system readiness encompasses biosensor, electronics packaging, and a mobile phone application for lactate monitoring by different users, contributing to cost-effectiveness.

Materials and Method

Lactate Oxidase (LOx), [3-(Methylamino) propyl] trimethoxysilane (MAPTS), Aquavion^® D98-25BS, sodiumL-lactate, potassium ferricyanide,  iron (III) chloride, hydrochloric acid, hydrogen peroxide, potassium chloride, nickel chloride, glucose, citric acid, acetic acid, urea, ascorbic acid, ethanol, isopropanol, Phosphate Buffered Saline (PBS) tablets were purchased from Sigma-Aldrich. Nafion polymer solution was purchased from ANR Technologies. Milli-Q Deionized (DI) water was used throughout. Concentration standards were prepared with sodiumL-lactate and PBS. The three-electrode cell consists of graphene foam Working Electrode (WE) and Counter Electrode (CE), and Ag/AgCl Reference Electrode (RE). Liquid drop-casting was done with the sciFlexarrayer S12 non-contact liquid dispensing system.

A Prussian Blue (PB) transuding layer was first synthesized on the chemical vapor deposited graphene layer (for the case of single layer graphene device) and graphene foam WE (for the case of three-electrode device). The precursor solution was prepared by mixing 1-7mM potassium ferricyanide, 1-7mM iron chloride, 0.05-0.8M potassium chloride, and 0.1-0.3M HCl. Then 0.5µL and 3µL of the precursor solution was drop-casted onto the graphene layer and graphene foam WE respectively, 0.5M of H₂O₂ was further added to the droplet. It is left to react for 30-60minutes to form PB, before rinsing it with DI water and drying it. The PB coated electrode is then annealed at 80°C for 30-60 minutes. A stabilizing solution of 1-3mM nickel chloride, 0.5-3mM potassium ferricyanide, 0.25-1M potassium chloride, and 0.1-0.3M HCl is prepared and 0.5µL and 3 µL was drop-casted onto the annealed Prussian blue layer of the graphene layer and graphene foam WE respectively and left for 30 minutes before rinsing it thoroughly with DI water and drying it. The nickel coated PB is then annealed at 80°C for 30 minutes. The addition of nickel improves the operational stability of the transducer during the detection of lactate.

LOx is suspended in DI water at 5 mg/ml and was added to a mix of 1-2% MAPTS, 10% Nafion in isopropanol to achieve a homogenous solution. The enzyme contained membrane mixture was dispensed over the transducing layer, 0.5 µL for the single layer graphene and 2µL for the graphene foam WE. The membrane is left to dry for 30 minutes before being cured at 4°C for 20 hours. For the sensor with larger concentration sensing capabilities, 2µL of Aquavion^®solution in isopropanol was dispensed after the membrane is dried and left to set at 4°C for 20 hours.

Following our functionalization protocols, we have developed two electrochemical sensors with distinct detection mechanisms. These sensors are designed to effectively cover an ultra-wide dynamic range of lactate concentrations, ranging from 100pM to 200mM.

The first sensor, characterized by higher sensitivity, comprises a two-probe device constructed from a single layer of graphene deposited using Chemical Vapor Deposition (CVD). Lactate detection occurs through conductivity measurements, wherein the current in our measurement channel increases via electron transfer. In Figure 1a, we show the optical image of one sensor with multiple measurement channels, overlaid with the two-probe measurement schematics. We also fabricated the three-electrode electrochemical sensors using our functionalization protocols to achieve wide concentration detection ranges (10 uM to 0.5mM and 1mM to 200 mM). In this approach, lactate detection occurs through chronoamperometry analysis, employing conventional working, counter, and reference electrodes.

For lactate detection with the sensors, amperometry measurements were conducted on the two-probe sensor with a Keithley 2450 source meter at 0.5V. For the three-electrode sensor: Amperometry measurements were conducted with Palmsens 4 at 0.0V applied voltage vs Ag/AgCl RE in batch mode (stirring at 200rpm) in PBS medium at pH 7.4. Calibration was done with the sodium L-lactate standards prepared for both types of sensors. Commercial sweat samples were measured with the three-electrode sensor, introduced in the batch mode with a dilution factor of 1,000 if needed. Operational stability of the three-electrode sensor was investigated at 0.1mM lactate in identical conditions. Selectivity tests of the three-electrode sensor was conducted similarly, with 0.2mM lactate as the control response.

L-lactate assay kit (Sigma-Aldrich, MAK329) was used to cross-validate readings from the sensor. To ensure consistency, it was intended for the same lactate standard solution to be used in building a standard curve for the assay kit and a linear calibration curve for the sensor. This standard solution was first tested against the sodium L-lactate provided by the kit, which was diluted to concentrations ranging from 0 to 2 mM, as per manufacturer’s protocol. The same was done for standard solution. The result showed that the concentration readings from both sodium L-lactate sources. Thus, the standard solution is identical to the sodium L-lactate provided by the kit.

Commercial sweat samples were first centrifuged at 10,000g for 10 minutes before being diluted by 20, 50, and 80 times with Milli-Q water. Absorbance value of the samples were detected at 565 nm with the spectrophotometer after reaction with the assay kit. After factoring in the dilution, the average is then calculated to obtain the actual concentration of the commercial samples.

Results and Discussion

The typical measurable range of detectable lactate concentrations of commercially available sensors ranges from 0.5 –25mM [20,21]. It is worth noting that the commercially available sensors present a narrow range of lactate concentration that can be detected [22]. However, these commercial sensors detect lactate only invasively through blood. Efforts have been undertaken to expand the detectable range, which has been successfully accomplished through the utilization of various experimental sensor configurations. Experimental non-contact sensing platforms based on lactate oxidase (LOx) have been able to detect lactate in a linear range of detectable concentrations ranging from 0.75-1000 μM (sweat, saliva) [23], 0.7-1500 μM (lactate stock solution) [24], 10-1000 μM (blood plasma) [25]. Likewise, the sensors based on lactate dehydrogenase (LDH) have been reported to have a linear range of detectable concentrations ranging 300-2000 μM (saliva and sweat) [26] 500-50,000 μM (human blood) [27]. However, broader detection ranges are needed when measuring across diverse lactate-containing fluids, including saliva, wine, blood, and sweat.

Figure 2a displays the two-probe response for a lactate concentration level of 100 pM. The current for a single channel was measured with an applied voltage bias of 500 mV. The conductivity measurement provides an instantaneous response upon the introduction of the lactate analyte and stabilizes within 100-150 s. The ratio of change in current over baseline current (ΔI/I₀) was used to calibrate the concentration levels of lactate (100pM – 100 µM, R² = 0.983), as shown in Figure 2b. Each analyte concentration was tested on a new sensor to fully explore the sensor's sensitivity and range.

Similarly, utilizing the three-electrode sensor, we continuously tested a single sensor across the concentration range from 0.01 mM to 1 mM as depicted in Figure 2c. The detection current levels are distinct from the previous experiment due to the different signal measurement schematic, and the measured signal via chronoamperometry analysis was directly used for the calibration of lactate analytes. Figure 2d shows the measured signals for various concentrations, depicting a clear linear response range from 0.01-0.5 mM, R² = 0.999.

Expanding on the three-electrode sensor, a negatively charged Aquavion^® layer was added and the sensor was tested across the concentration range 0.1mM to 200mM. Figure 2e shows the raw signal obtained over chronoamperometry, and the data was analyzed for the linear calibration curve of 1mM to 200mM, R² = 0.968 of lactate concentrations as shown in Figure 2f.

Figure 1. Sensor layout and measurement schematics. a) Optical microscope image of a single layer graphenedevice, overlaying with the measurement schematics for a 2-probe setup. The device consists of multiple activation channels that could be functionalized individually; b) Measurement schematics for a conventional three-electrode electrochemical sensor via chronoamperometry analysis. WE = working electrode, RE = reference electrode, CE = counter electrode and c) Precise and accurate liquid drop-casting with the automated sciFlexarrayer S12 which enables targeted auto dispensing of multi-chemicals down to volume of ~100 pL.

Figure 2. Ultra-wide range detection of lactate with graphene-based bio-sensors. a) Typical response curve, current vs time plot (I-t) for a single layer graphene device, showing a cleardetection signal upon the introduction of a standard lactate solution (1 nM); b) The current change in a) is used to calibrate the concentration levels of lactate from 100 pM to 100 μM; c)Response curve, current vs time plot (I-t) for a three-electrode device, showing current changes upon introduction of standard lactate solution; c inset) Response curve, current vs time plot(I-t) ranging 0.01 mM to 1 mM of lactate; d) Linear calibration curve of the three-electrode sensor, covering a lactate range from 10 μM to 0.5 mM; d inset) Linear calibration curve covering 0.01 mM to 0.5 mM of lactate; e) Response curve, current vs time plot (I-t) for a three-electrode device, showing current changes upon introduction of concentrated standard lactate solution; f) Linear calibration curve of the three-electrode sensor, covering a lactate range from 1 mM to 200 mM; g) A response curve vs time showing the selectivity tests of the three-electrode sensor; h) Operational stability (at 0.1 mM lactate) over 10 hours; i) A plot of the measured value of lactate solutions with our sensor against the nominal standard concentration (3 data set for each concentration); j) Standard deviation in percentage of the 3 measurements at each concentration.

We have also tested the sensor’s ability to detect only the analyte (lactate) without being significantly influenced or triggered by other substances present in the sample minimizing false positives from interfering substances. This characteristic is crucial for ensuring the sensor's reliability and usefulness in various applications, such as medical diagnostics, environmental monitoring, and food safety testing. To this end, different solutions were added to the sensor and the current changes were measured. The absence of well-defined, high-intensity peaks for all analytes except lactate confirms the sensor's high selectivity towards lactate.

Lactate oxidase achieves this high specificity with its mechanism: Initially combined with flavin mononucleotide (FMN) as a cofactor; it oxidizes lactate to form an intermediate complex with FMN in the first reductive half step while producing pyruvate, then re-oxidizing its FMN in the second oxidative half step to its original lactate oxidase and FMN state while releasing hydrogen peroxide as a byproduct [28]. This complex of lactate oxidase and FMN contains a permutation of amino acids, functional groups and space in the active site that specifically requires the orientation of a methyl, hydroxyl and carboxyl functional groups found in L-lactate to be bound [29], thereby providing the specificity noted here. The hydrogen peroxide then undergoes a redox reaction, catalyzed by the PB. The electron released in the process is then detectable. Graphene’shigh electron mobility enable this highly accurate, precise and rapid detection, seen in the sensitivity of the sensors, with notable signal changes. In addition, the negatively charged polymer Aquavion^® produces a repelling effect on the negatively charged lactate molecule, forming a pseudo-competitive inhibitory effect, which in turn, only allows lactate to be bound by the lactate oxidase when an excessive amount of lactate is present. This thereby shifts the range of the lactate concentrations detectable in the sample Table 1.

Table 1. Comparison of sweat lactate concentration readings from L-lactate assay kit and three-electrode sensor. Analysis with Paired Sample t-Test using OriginPro showed that difference in concentration readings from sensor and assay kit is negligible (t = -0.53, df = 10, p = 0.61).

The sensitivity was calculated to be 0.0850 ± 0.00456 Δ·M^-1, 6.668 ± 0.0537µA·mM^-1, 1.278 ± 0.116µA·mM^-1 for the single layer graphene sensor, the three-electrode sensor, and the larger concentration three-electrode variant respectively. In addition, the Limit of Detection (LOD) and Limit of Quantification (LOQ) are calculated to be 1pM and 100pM for the single layer graphene sensor; 1μM and 0.01mM for the three-electrode sensor; and 0.01mM and 1mMfor the higher concentration three-electrode variant respectively. In particular, the sensitivity of the three-electrode sensor is higher than those of previously reported screen-printed sensors. This is a significant improvement over the Limits of Detection (LOD) reported in current commercially available sensors that range from 0.3 to 0.7 mM [22] as the LOD denotes the minimum concentration of lactate reliably detectable by the sensor (smallest measurement above background noise). It is however of note that LOQ is a more practical measure, that being of the minimum concentration reliably quantifiable for analytical use. It is noteworthy that the high signal strength and stability observed at all concentrations suggest potential not only a high sensitivity for detecting lower lactate concentrations but also a high potential for reproducibility.

Conclusion

In conclusion, we have demonstrated functionalization protocols for fabricating a highly sensitive lactate graphene sensor presenting a significant advancement in biosensor technology. By deploying two distinct sensor detection mechanisms, we have achieved a wide coverage of lactate concentration levels, ranging from an ultra-low 100 pM to a high concentration of 200 mM, enabling an ultra-wide detection range. This broad dynamic range holds substantial promise for applications in healthcare monitoring and point-of-care diagnostics, facilitating early intervention and personalized treatment strategies. The versatility and precision of our sensor design underscore its potential to medical diagnostics, offering enhanced sensitivity and accuracy for improved patient care and disease management.

Acknowledgements

This research was carried out at the Centre for Advanced 2D Materials (CA2DM), funded by the National Research Foundation, Prime Minister's Office, Singapore, under its Medium-Sized Centre Programme and Central GAP.

Conflict of Interest

The authors declare no conflict of interest.

Author Contributions

S.T., G.K.W.K. and A.H.C.N. conceptualized the project. S.T., G.C.F.S. and J.J.Q.N.  Performed the measurements. S.T., G.C.F.S., J.J.Q.N. and G.K.W.K. performed the analysis and discussion. S.T., G.K.W.K. and V.R. led the manuscript writing with the assistance of all co-authors. A.P.L. managed the project. A.H.C.N. supervised the project.

References

1. Rosenstein PG, Tennent‐Brown BS, Hughes D (2018) Clinical use of plasma lactate concentration. Part 2: Prognostic and diagnostic utility and the clinical management of hyperlactatemia. J Vet Emerg Crit Care 28: 106-121. [Crossref]
2. Toffaletti JG (1991) Blood Lactate: Biochemistry, Laboratory Methods, and Clinical Interpretation. Crit RevClin Lab Sci 28: 253-268. [Crossref]
3. De Backer D, Creteur J, Zhang H, Norrenberg M, Vincent JL (1997) Lactate Production by the Lungs in Acute Lung Injury. Am J Respir Crit Care Med 156: 1099-1104. [Crossref]
4. Lazzeri C, Valente S, Chiostri M, Gensini GF (2015) Clinical significance of Lactate in acute cardiac patients. World J Cardiol 7: 483-489. [Crossref]
5. Sonkodi B (2022) Should We Void Lactate in the Pathophysiology of Delayed Onset Muscle Soreness? Not So Fast! Let’s See a Neurocentric View! Metabolites 12: 857. [Crossref]
6. Bonaventura JM, Sharpe K, Knight E, Fuller KL, Tanner RK, et al. (2015) Reliability and accuracy of six hand-held blood lactate analysers. J Sports Sci Med 14: 203-214. [Crossref]
7. San-Millán I (2019) Blood Biomarkers in Sports Medicine and Performance and the Future of Metabolomics.Methods Mol Biol 1978: 431-446. [Crossref]
8. Derbyshire PJ, Barr H, Davis F, Higson SPJ (2012) Lactate in human sweat: a critical review of research to the present day. J Physiol Sci 62: 429-440. [Crossref]
9. Villena Gonzales W, Mobashsher A, Abbosh A (2019) The Progress of Glucose Monitoring—A Review of Invasive to Minimally and Non-Invasive Techniques, Devices and Sensors. Sensors 19: 800. [Crossref]
10. Harb F, Azar WS, Ghadieh HE, Njeim R, Tawk Y, et al. (2022) Future Developments in Invasive and Non-invasive Diabetes Monitoring. In: Sadasivuni KK, Cabibihan J-J, A M Al-Ali AK, Malik RA, editors. Advanced Bioscience and Biosystems for Detection and Management of Diabetes. Vol. 13. Cham: Springer International Publishing 293-313.
11. Seki Y, Nakashima D, Shiraishi Y, Ryuzaki T, Ikura H, et al. (2021) A novel device for detecting anaerobic threshold using sweat lactate during exercise. Sci Rep 11: 4929. [Crossref]
12. Jia W, Bandodkar AJ, Valdés-Ramírez G, Windmiller JR, Yang Z, et al. (2013) Electrochemical Tattoo Biosensors for Real-Time Noninvasive Lactate Monitoring in Human Perspiration. Anal Chem 85: 6553-6560. [Crossref]
13. Khodagholy D, Curto VF, Fraser KJ, Gurfinkel M, Byrne R, et al. (2012) Organic electrochemical transistor incorporating an ionogel as a solid state electrolyte for lactate sensing. Journal of Materials Chemistry 22: 4440.
14. Gao W, Emaminejad S, Nyein HYY, Challa S, Chen K, et al. (2024) Fully integrated wearable sensor arrays for multiplexed in situ perspiration analysis. Nature 529: 509-514. [Crossref]
15. Currano LJ, Sage FC, Hagedon M, Hamilton L, Patrone J, et al. (2018) Wearable Sensor System for Detection of Lactate in Sweat. Sci Rep 8: 15890. [Crossref]
16. Gao F, Liu C, Zhang L, Liu T, Wang Z, et al. (2023) Wearable and flexible electrochemical sensors for sweat analysis: a review. Microsyst Nanoeng 9: 1. [Crossref]
17. Terse-Thakoor T, Punjiya M, Matharu Z, Lyu B, Ahmad M, et al. (2020) Thread-based multiplexed sensor patch for real-time sweat monitoring. npj Flexible Electronics 4: 18. [Crossref]
18. Ramachandran B, Liao Y-C (2022) Microfluidic wearable electrochemical sweat sensors for health monitoring. Biomicrofluidics 16: 051501. [Crossref]
19. Heikenfeld J, Jajack A, Rogers J, Gutruf P, Tian L, et al. (2018) Wearable sensors: modalities, challenges, and prospects. Lab Chip 18: 217-248. [Crossref]
20. Vivachek^TM Lactate Analyzer. [accessed 2024 Apr 8].
21. Lactate Scout Sport Lactate Analyzer | EKF Diagnostics. [accessed 2024 Apr 8].
22. Kucherenko IS, Topolnikova YaV, Soldatkin OO (2019) Advances in the biosensors for lactate and pyruvate detection for medical applications: A review. TrAC Trends in Analytical Chemistry 110: 160-172.
23. Cunha-Silva H, Arcos-Martinez MJ (2018) Dual range lactate oxidase-based screen printed amperometric biosensor for analysis of lactate in diversified samples. Talanta 188: 779-787. [Crossref]
24. Romero MR, Garay F, Baruzzi AM (2008) Design and optimization of a lactate amperometric biosensor based on lactate oxidase cross-linked with polymeric matrixes. Sensors and Actuators B: Chemical 131: 590-595.
25. Haghighi B, Bozorgzadeh S (2011) Fabrication of a highly sensitive electrochemiluminescence lactate biosensor using ZnO nanoparticles decorated multiwalled carbon nanotubes. Talanta 85: 2189-2193. [Crossref]
26. Smutok O, Karkovska M, Serkiz R, Vus В, Čenas N, et al. (2017) A novel mediatorless biosensor based on flavocytochrome b 2 immobilized onto gold nanoclusters for non-invasive L -lactate analysis of human liquids. Sensors and Actuators B: Chemical 250: 469-475.
27. Batra B, Narwal V, Pundir CS (2016) An amperometric lactate biosensor based on lactate dehydrogenase immobilized onto graphene oxide nanoparticles‐modified pencil graphite electrode. Engineering in Life Sciences 16: 786-794.
28. Hiraka K, Yoshida H, Tsugawa W, Asano R, La Belle JT, et al. (2022) Structure of lactate oxidase from Enterococcus hirae revealed new aspects of active site loop function: Product‐inhibition mechanism and oxygen gatekeeper. Protein Sci 31: e4434. [Crossref]
29. Umena Y, Yorita K, Matsuoka T, Kita A, Fukui K, et al. (2006) The crystal structure of l-lactate oxidase from Aerococcus viridans at 2.1 Å resolution reveals the mechanism of strict substrate recognition. Biochem Biophys Res Commun 350: 249-256. [Crossref]