Open Access
Add to BibSonomy
Abstract
Alanine aminotransferase (ALT), a critical component of human blood, is inextricably associated with liver injury. The current study develops a novel biosensor based on the localized surface plasmon resonance (LSPR) principle for the detection of ALT analytes at concentrations ranging from 0 to 1000 Units per liter (U/L). According to the authors' knowledge, this is the first time an optical fiber structure with a taper-in-taper structure has been developed for biosensing applications. It is fabricated using the three-electrode semi-vacuum taper technique and is characterized using a combiner manufacturing system. Gold nanoparticles (AuNPs), molybdenum disulfide nanoparticles (MoS2-NPs), and cerium oxide nanoparticles (CeO2-NPs) are immobilized on the sensing region to improve the sensing performance. Prior to application, these nanoparticles are characterized using a high-resolution transmission electron microscope (HR-TEM) and a UV-Visible spectrophotometer. AuNPs promote the LSPR phenomenon, whereas MoS2-NPs/CeO2-NPs contribute to the sensor probe's biocompatibility and stability. Following that, the probe surface was functionalized with glutamate oxidase (GluOx) to improve selectivity. The probe demonstrated an excellent linear relationship with the subsequent assay's ALT concentration. Additionally, the probe's performance characteristics such as reusability, reproducibility, stability, and selectivity are evaluated in order to determine its clinical utility in diagnosing liver injury.
© 2021 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Alanine aminotransferase (ALT), a necessary liver enzyme, is a type of serum transaminase whose activity must be determined in order to diagnose and estimate liver disease. Clinical trials and experimental testing have revealed that when certain metabolic disorders such as obesity, high-fat diabetes, and other symptoms manifest in the body, a mild increase in ALT levels may occur. Additionally, ALT activity in a variety of muscle diseases such as viral hepatitis and muscular dystrophy results in a relatively elevated ALT levels [1]. To increase the reliability of screening for overall health status during routine health checking and to help understand the role of the ALT enzyme as an analytic assessment tool, researchers have spent years attempting to determine the clinical factors that affect ALT levels [2]. ALT is a necessary component of the tricarboxylic acid (TCA) cycle, in that the ALT enzyme catalyzes the transfer of the amino group from L-alanine to a-ketoglutaric acid to produce L-glutamate and pyruvic acid [3,4]. According to articles [2,5,6], ALT is found in a variety of organs and tissues, including adipose tissue, the brain, the liver, and the intestines. Among them, the activity of ALT in the liver is significantly greater than that in other human organs. When the liver is severely damaged, then ALT from injured liver's cells enters the serum, resulting in a massive increase in Alanine transaminase activity in the serum. The ratio of enzyme activity in the liver to that in the serum is approximately 3000:1. Meanwhile, the normal activity level of the ALT enzyme in serum is between 5 to 40 U/L [7]. As a barometer of the health of the liver, abnormal levels of ALT enzyme in the blood (tens of folds of the normal range from 250-1,400 U/L) can indicate damage to the organ [8]. Thus, whether the liver is seriously damaged can be determined indirectly from clinical trials by examining the level of ALT activity in serum samples [9]. Despite of these challenges, scientists and research organizations have made tremendous strides in recent decades in the research of clinical monitoring techniques for patients with liver diseases [1,2]. For example, the key to colorimetric analysis [10] is to calculate and draw a set of calibration curves for the amount of product produced and substrate consumed during the enzyme reaction, and then to calculate the concentration of the analyte using the curves. However, the specificity may be inadequate in the case of an icteric, lipemic serum sample. As a result, complex reaction environments, such as those described in [11,12], are unsuitable for point-of-care application. The techniques discussed previously, such as colorimetry, chemiluminescence, ultraviolet absorption, fluorescence, spectrophotometry, and electrothermal have some inherent difficulties and technical shortcomings [7,13,14]. In recent years, optical techniques such as plasmonic and particularly localized plasmons have garnered attention for the detection of various biosystems due to the benefits of tailored optical fiber. In the field of biosensing, it has been reported that noble metal nanoparticles (NPs) have a wide range of advantages such as high surface-area-to-volume ratio [15], catalytic characteristic [16], excellent biocompatibility [17], and good surface chemical activity [18]. There has been a significant increase in interest in the localized surface plasmon resonance (LSPR) which involves interactions of metallic spherical NPs of radius R (such that R is below the diffraction limit) immersed in a dielectric analyte with the incident electromagnetic radiation. The coherent electronic oscillations are set off on the surface of the NPs, termed as localized surface plasmons that oscillate at the corresponding LSPR frequency [19,20]. LSPR biosensors detect the presence of analytes in a small target solution by monitoring the refractive index change (RI). To our knowledge, LSPR-based biosensors for ALT biomolecule detection have the potential to pave the way for the development of a highly sensitive, label-free, real-time, and rapid detection method for ALT biomolecules.
Molybdenum disulfide nanoparticles (MoS2-NPs) and cerium oxide nanoparticles (CeO2-NPs) are two commonly used heavy metal nanoparticles that are biocompatible, nontoxic, and have a large specific surface area, a high conductivity, and excellent electron transfer properties [21,22]. CeO2-NPs have been widely used as a single nanoparticle modifier in electrochemical sensing to date due to their excellent conductivity, chemical stability, high specific heat capacity, and high electron transfer properties. It has been used as a critical multifunctional auxiliary material in a variety of biosensors in recent years [23]. CeO2-NPs, for example, can be used in conjunction with indium–tin oxide to enhance enzyme adsorption [24]. Dong et al. used CeO2-NPs as doping particles in electrochemical biosensors to increase their sensitivity and stability [25]. Nnguyet et al. modified the electrode with cerium oxide nanoparticles (CeO2-NPs) and polypyrene nanoparticles (PPy-NPs) to increase the sensor's sensitivity for salmonella detection. The CeO2-NPs significantly improved the stability of covalent bonds between single-stranded DNA (ssDNA) sequences and microelectrodes [26]. As a result, CeO2-NPs are also very promising as biosensing materials.
The purpose of this work is to design and fabricate a taper-in-taper structure sensing probes with single mode fiber (SMF). To increase the LSPR effect and thus the sensor sensitivity, NPs such as gold nanoparticles (AuNPs), MoS2-NPs, and CeO2-NPs are immobilized on the sensing region of the probe. Simultaneously, the glutamate oxidase enzyme is functionalized over an NPs-immobilized fiber structure, enhancing the specificity of the sensor. Prior to the analyte test, the prepared NPs and probes were characterized and examined microscopically. Finally, the sensing performance of the sensor is evaluated experimentally, and the results indicate that the sensor has exceptional potential for diagnosing liver injury.
2. Experimental section
2.1 Materials
To avoid interference with the output waveform caused by increasing the transmission mode, sensor probes with a special taper-in-taper structure were fabricated using conventional single-mode optical fiber (SMF, 9 μm, 125 μm, EB-link Technologies Co, Shenzhen). Chloride-trihydrate (HAuCl4) and sodium citrate (C6H5Na3O7) are primarily used in the Turkevich protocol for the synthesis of AuNPs. N-Methyl-2-pyrrolidone (NMP) is an excellent solvent for dissolving MoS2-NPs and is therefore well suited for use in this experiment. As an activator, 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC) and N-Hydroxysuccinimide (NHS) are used to increase the efficiency of the coupling reaction with 11-Mercaptoundecanoic acid (MUA). Glutamate oxidase (GluOx) from Streptomyces (39346-34-4, Sigma-Aldrich, Shanghai) are primarily used to functionalize the surface of optical fiber probe to add the specificity property. The Alanine aminotransferase (ALT), α-ketoglutarate, and L-alanine were purchased from Sigma-Aldrich, Shanghai. Acetone was used as a ketone cleaning agent to remove a variety of organic compounds from the surface of the optical fiber. For drying the fiber probes, nitrogen (N2, a type of asphyxiant gas) was used. Other analytical-grade reagents such as ethanol, ascorbic acid, potassium hydroxide, glucose, concentrated sulfuric acid, and concentrated hydrochloric acid were purchased locally.
2.2 Instrument and measurement
The following high-precision testing instruments were used in this study to obtain more precise experimental results. To begin, the 3SAE combiner manufacturing system (CMS, USA) was used to fabricate and characterize well-structured fibers that would later be functionalized with sensing probes. An ultraviolet-visible (UV-Vis) spectrophotometer (U-3310, HITACHI, Japan) was used to determine the absorbance spectrum of the prepared NPs solution. The microscopic distribution of the NPs in the synthesized solution was confirmed using a high-resolution transmission electron microscope (HR-TEM, Talos L120C, Thermo Fisher Scientific). A scanning electron microscope (SEM, Gemini Carl Zeiss microscopy) was used to observe the NPs-coated layer on the sensing probe's surface. Prior to the final measurement, the fiber probe was characterized using the instruments listed above. Furthermore, the detection device makes use of a light source and spectrometer.
2.3 Sensing mechanism of the probe
The core of a conventional optical fiber transmits light via total internal reflection (TIR), and at the interface between the core and the cladding, a small amount of evanescent waves (EWs) is generated. EWs occur as a result of electromagnetic waves’ TIR as they travel from the core to the low RI cladding, penetrating slightly to several wavelengths. The generated EWs propagate parallel to the interface between the core and cladding as a surface wave, and their amplitude decreases exponentially with the vertical interface direction. Due to the fact that the fractional power of EWs is extremely sensitive to environmental changes, the sensor can be designed and developed using this principle [27,28].
When fibers have a tapered structure, as the diameter of the core and cladding decreases, more light fields are excited and high EWs are generated on the cladding's outer surface. Simultaneously, the waist sensing region has a significantly greater EWs power than the conventional sensing region [29]. Because changes in the propagation constant and the excitation of additional higher-order modes can improve sensor sensitivity, that is closely related to the unique structure of the transition region [30–32]. Thus, in comparison to the conventional fiber structure, a novel taper-in-taper optical fiber structure was developed and evaluated for use in additional sensing applications. Due to the addition of tapering processing to the conventional taper structure, the novel shape of the taper region is better suited for exciting the field in the sensing region and producing more high-power EWs. Figure 1 illustrates the taper-in-taper structure. After twice-time tapering toward the structured fiber, the diameters of the core and cladding are uniformly reduced in proportion.
When the incident light frequency is equal to the oscillation frequency of the surface wave, the plasma wave is generated only in the vicinity of the sensing region. LSPR is the term used to describe the phenomenon that results in the collective oscillations of conducting electrons within a metal particle. The LSPR effect causes the interaction of NPs to alter the distribution of the evanescent field by absorbing the energy of the partial EWs and scattering the light outside the geometric cross-section. Thus, the spectral shift of LSPR can be determined by measuring the transmission or reflection intensity. The absorption of energy is the primary factor for NPs smaller than 20 nm, and the comfortable field increases dramatically when the diameter is less than 10 nm, and that's why the experiment used NPs with a diameter of approximately 10 nm [33]. Additionally, the LSPR effect's strong and highly localized electromagnetic field makes NPs on the surface extremely sensitive to small changes in the local RI [34]. Thus, when the analyte in the sample is combined specifically with the functional enzyme, a change in the RI near the NPs occurs, and the analyte concentration can be determined. Besides, when noble metal NPs are fixed on the surface of the sensor, they are sensitive to subtle changes in the surrounding environment (mainly small changes in RI), altering the absorption frequency and intensity of noble metal surface plasmas, as well as the LSPR signal received by the sensor. Thus, it is intuitionistic to see the shift of resonance peak wavelength. In addition, more EWs energy on the surface of sensing region will be favorable to induce stronger LSPR effect, and a good portion of this energy will eventually be coupled back to the fiber core in the form of other energy, therefore, the performance and quality of the sensor can not be evaluated simply by the change of light intensity.
To increase the enzyme's contact point and thus the detection sensitivity, MoS2 and CeO2 NPs were used to immobilize over the AuNPs nano-coating. This is because MoS2-NPs and CeO2-NPs are both common heavy metal NPs that are biocompatible, nontoxic, have a large specific surface area, a high conductivity, and excellent electron transfer properties [21,22]. As an effective bridge between the enzyme molecules and AuNPs, MoS2 can effectively improve the surface load capacity of macromolecules, meanwhile, unbonded sulfur of the nanomoles can be combined with precious metals to form a highly coupled grid capacitor [35]. In addition, when AuNPs combined with MoS2-NPs, the amount of GluOx molecules adhesion increased with the increase of the total surface area, that helped to improve the sensitivity of the sensor. However, despite of the fact that the MoS2-NPs layer has a high surface-to-volume ratio, the interfacial performance of the total immobilization is poor, that will negatively impact the adhesion of the enzyme to the surface and, consequently, the by-product of enzyme coating. As a result, CeO2-NPs is chosen as a modifier to improve the overall optical properties of the probe and to make the structure of the fixed layer on the surface of the probe more stable. CeO2-NPs is a nanoparticle that has been designed to enhance the overall optical properties of the probe [36,37].
2.4 Fabrication of sensor probe
The combiner manufacturing system (CMS) is the primary tool used in this work to fabricate the novel SMF-based taper-in-taper structure described in this article. The proposed structure optical fiber is heated using a three-electrode thermal stable plasma heating system, that is a unique technology developed by the CMS machine for the fabrication process. The vacuum saturation and the power of the discharge point between the electrodes are both controlled by a programme to ensure that the proper heating conditions are achieved between the electrodes. Because the plasma is thermally stable and the heating temperature is not affected by objective factors such as electrode wear and ageing, the heating technology is far more advanced than the traditional arc heating technology used in most industrial applications. Continuous testing may enable a well-optimized parameter within CMS to achieve twice-fixe-point SMF tapering. The following is a summary of the CMS tapering treatment procedure. To begin, the conventional SMF was stripped of its coating layer and then carefully clamped and encapsulated in CMS. Using the principle of gas pressure, the valve system equipment in CMS generated a semi-vacuum environment around the SMF. When the ambient vacuum reaches a predetermined value of 3, the three electrodes discharge in perfect harmony, forming an even plasma field on the fiber surface. This operating principle preserves the optical fiber's uniform heating point, that is more conducive to the formation of a symmetrical taper structure. Once the heating power and softening temperature are set, the two-sided mechanical motor is used to force the clamps to lengthen the optical fiber by a minimum of approximately 3 μm before forming a new heating point and stretching it again. After two minutes of extremely rapid repetition of the preceding heating while stretching procedure, an intact taper structure is formed. Figure 2(a) illustrates the working schematic during the initial tapering treatment. Additionally, it is vital to consider that the starting and waist heating powers cannot be set too far apart. The diameter scanning result is obtained using CMS after fabricating a complete single-tapered structure of SMF with an 80 μm waist diameter. Following that, as illustrated in Fig. 2(b), fine-tune the starting position of the second taper operation based on the results of the previous scan.
Fig. 2. (a) Initial tapering process view inside the CMS instrument, (b) second tapering operation based on the results of the previous scan, and (c) taper-in-taper fiber structure fabrication process.
While the second tapering procedure is similar to the first, certain parameters are different. Numerous optimizations have been made to the taper programme, and now it is possible to develop taper-in-taper structure fibers with excellent reconstruction simply by calling the corresponding programme. The simplified manufacturing flow diagram in Fig. 2(c) illustrates the manufacturing process for an SMF-based taper-in-taper structure, starting with a common fiber and ending with a design structure.
2.5 Nanoparticle synthesis process
The conventional Turkevich protocol was primarily used to synthesize the 10 nm AuNPs. The steps in the synthesis are as follows: to begin, aqua regia-washed bottle was used to contain 15 mL of HAuCl4 aqueous solution (150 μL, 100 mM HAuCl4 in 14.85 mL DI water), that was then heated to boiling (approximately 110 °C). Following that, 1.8 mL of 38.8 mM sodium citrate solution was added and heated to 110 °C for 5 minutes. Finally, turn off the heat and left the stirring function running for about 10 minutes. At this point, the synthesis of the AuNPs solvents was completed. The solution of the synthesized AuNPs demonstrated excellent stability, with the ability to be stored in a cool, dry environment for three months without aggregation [38,39].
A liquid-phase exfoliation protocol was used to produce the MoS2-NPs solution. Simply put, 30 mg MoS2 powder was dissolved in 10 mL of NMP solution and placed in an ultrasonic bath machine for ultrasonic pulverization. After 15 minutes, the suspension was transferred to a sonicator probe for an additional 10 minutes of stripping treatment. After centrifuging the treatment solution at a rate of 4000 rpm for 1 hour, the required synthesized solution is obtained for subsequent use.
The steps for the synthesis of CeO2-NPs include placing 0.1 g of cerium nitrate hexahydrate in a bottle, adding 49 mL DI water, and then slowly dropping 1 mL 30% H2O2 solution. After the steps above, the color of the solution will gradually be turned light yellow due to oxidation and reduction. Finally, the prepared solution will be left for 3 weeks, and the color of solution will gradually become colorless.
2.6 Nanoparticle coating and enzyme functionalization
To proceed, the acetone solution is used for 20 minutes to clean the surface of the fiber probes, removing organic matter, and smoothing the cladding surface. Secondly, the proposed probes are hydrolyzed by dipping the sensing regions of cleaned fibers in the Piranha solution (3 volumes of 30% H2O2 and 7 volumes of concentrated H2SO4) for 30 minutes. Thirdly, completely clean the fiber probes with DI water and place them in an oven set to 70 °C for approximately 20 minutes. Fourthly, for 12 hours, dip the dried sensing regions of fibers in ethanolic 1% MPTMS solution. MPTMS acts as a coupling reagent in this case, facilitating AuNPs adhesion to the useful thiol-functionalized monolayer on the fiber surface [27]. Fifthly, the probes were rinsed in ethanol and then dried with nitrogen gas to remove any non - fixed MPTMS monomers on the fiber probes surface. Through a simple and economical dip coating technique, the cleaned prepared fiber probes are dipped in AuNPs aqueous solution to immobilize the AuNPs for 48 hours. Following that, rinse the probes with ethanol and dry them with nitrogen gas to remove the unbounded AuNPs.
Similarly, to the dip coating procedure described previously, the AuNPs-coated sensing region was immersed for 20 seconds in a 10 mL MoS2-NPs solution and dried for 2 minutes at 50 °C. The MoS2 dip coating process was repeated eight times to ensure that the MoS2-NPs layer covers the surface of the fiber probes uniformly. Following that, placed the proposed fibers in a constant temperature oven set to 50 °C for 2 hours to increase the robustness of the MoS2-NPs immobilized probes [38].
Thereafter, sensing region of fiber probes was completely immersed in a 10 mL of CeO2-NPs solution for 5 minutes at room temperature. The next step is quenching, that involves simply putting the fiber in a constant temperature oven and heating it to 70°C for 30 minutes. Then, twice more, repeat the dipping and quenching steps. Finally, fiber probes with uniform AuNPs, MoS2-NPs, and CeO2-NPs distribution layers were obtained as shown in Fig. 3.
2.7 Enzyme functionalization
The prepared NPs-coating probes were immersed in 10 mL of 0.5 mM MUA ethanolic solution for 5 hours. MUA is a type of self-assembled monolayer (SAM) that contains a high concentration of alkane chains and carboxyl groups, allowing for subsequent enzyme functionalization [40,41]. As a result, the fiber probes were surrounded by a layer of abundant alkane chain and carboxyl group immobilized on the NPs surface. To immobilize the specific enzyme, the fiber probes would be immersed in a 5 mL solution of EDC (200 mM) and NHS (50 mM). As a non-toxic, biocompatibility crosslinking reagent, EDC can induce the cross-linking between the amino groups of enzymes and the carboxyl groups on the surface of the fixing layer, forming a collagen scaffold with excellent cellular compatibility [42]. However, the cross-linking effect of EDC is most effective at pH 4.5, the amount of that reagent can be increased to ensure the efficiency of enzyme immobilization in 1×PBS (pH 7.4). In general, EDC is used in combination with the NHS at neutral condition, they are coupled to carboxyl groups to form a stable NHS lipid that can be effectively jointed to an enzyme. In other words, the NHS can act as an enhance reagent that can available to improve its coupling [43,44]. After about 30 minutes of soaking, the carboxyl group on the surface was activated [45]. The NHS ester group reacted with the functionalized enzyme after 12 hours of immersion in 100 U/L (0.1 mg of 5 U/ mg solid dissolved in 5 mL of PBS) of GluOx solution, forming a uniform coating of GluOx on the sensing surface [46]. After the preceding steps were completed, enzyme functionalized fiber probes were obtained. As illustrated in Fig. 4, the enzyme functionalization process was carried out at room temperature.
Fig. 4. Schematic of enzyme functionalization over nanoparticles-immobilized optical fiber structure.
2.8 Preparation of analytes
The ALT solution contains various types of concentrations ranging from 0 to 550 ng/mL, that corresponds to an ALT specific activity of approximately 0–1000 U/L. Two steps are required prior to preparing the analytes. In the first step, 1 mM α-ketoglutarate and 100 mM L -alanine were used as the ALT substrate, that were stored at 35 °C. The second step is to prepare ten ALT solution at various concentrations (0, 10, 25, 50, 75, 100, 250, 500, 750, 1000 U/L). Then, the procedure for preparing the analytes can be summarized as follows. First, prepare a stock solution by dissolving an ALT (10 mg of 80 U/mg) in 1 mL of PBS solution (pH = 7.4). Then, 10 mL of a 1 U/L ALT stock solution was diluted with 1 mL of a 10 U/L ALT stock solution. To prepare a 5 mL solution of each concentration, PBS solution can be used to modify lower concentration samples such as 750, 500, 250, 100, 75, 50, 25, and 10 U/L from higher concentration samples. Then, to perform the reaction prior to performing the final measurement, one substrate tube was taken, and one type of ALT enzyme was added. The order of the measurements is from lower to higher ALT concentrations.
To ensure measurement accuracy, all experiments were conducted at the same temperature (room temperature) and the same reaction time (3 min after adding ALT). The catalytic reaction of ALT enzyme with co-substrate is shown in Fig. 5. The different concentration of L-glutamate materials produced by the reaction can be combined with the oxidation of the immobilized GluOx enzyme on the surface of probe that is used as the target of detection to measure the activity concentration of ALT enzyme.
2.9 Experimental setup
A halogen light source emits a specific amount of light that is propagated through the core to the sensing probe. Following that, the probe was immersed in the sample solution on the reaction platform. The resulting analyte will specifically bind to the immobilized enzyme on the surface, causing a change in the RI near the metal immobilization layer. The transmitted signals are then recorded with a spectrometer, and the resulting data is displayed on a laptop. Figure 6 illustrates the sensing experimental setup.
3. Results and discussions
3.1 Optimization of an optical fiber sensor probe
In traditional tapered structures, the waist length, waist radius, and tapered length are all critical parameters for sensitivity. The tapered length influences the number of peaks and valleys in the transmission spectrum, whereas the smaller waist radius increases optical fiber sensors’ leakage and sensitivity.
Meanwhile, the novel structure of the taper region alters the propagation constant and excites additional higher-order modes, increasing the sensitivity of the sensor. As a result, this work developed a taper-in-taper structure to increase the sensor's RI sensitivity. The SMF has been tapered twice, with the first waist region having a diameter double that of the second. The length of the taper transition region and the waist region of the conventional SMF are 2 mm and 8 mm, respectively, following the initial tapered fabrication protocol. Additionally, a second tapered fabrication is performed based on the taper structure's waist region. Following that, a taper-in-taper SMF with an 8-mm sensing region and a 40 μm diameter was obtained. The diameter scan results of three distinct proposed fibers fabricated using the same process are shown in Figs. 7(a) and 7(b). As can be seen, the optical fiber structures obtained have excellent structural characteristics, highly consistent curves, and high repeatability.
Fig. 7. Scan results of three distinct fabricated proposed fiber structures (a) 80 μm tapered fiber, (b) taper-in-taper fiber structure (80-40 μm).
3.2 Characterization of nanoparticles
The wavelength of their absorption peaks and the morphology of microscopic particles in the synthesized solution are used to identify NPs. AuNPs have an absorbance peak at 519 nm, as shown in Fig. 8(a). The HR-TEM was then used to capture images of the AuNPs distribution in the sample solution. As illustrated in Fig. 8(d), the synthesized NPs are highly uniform in shape and exhibit no agglomeration. To confirm the successful synthesis of nanomaterials for proposed work, a similar method was used to measure the wavelength of the peak absorbance of the other two NPs solutions. The absorbance wavelength peak of MoS2-NPs and CeO2-NPs solutions was determined by UV-Vis spectrophotometer at 330 nm and 253 nm, as shown in Fig. 8(b) and Fig. 8(c), respectively. At the same time, HR-TEM was used to characterize the morphology of MoS2 and CeO2-NPs. Large-scale micro-MoS2-NPs have been successfully synthesized as shown in Fig. 8(e). The TEM image of CeO2-NPs is shown in Fig. 8(f).
Fig. 8. Absorbance spectra of (a) AuNPs, (b) MoS2-NPs, (c) CeO2-NPs and HR-TEM images of (d) AuNPs, (e) MoS2-NPs, (f) CeO2-NPs
3.3 Characterization of nanoparticle-immobilized taper-in-taper structure
SEM was used for high-resolution micro-topography analysis of sensor probe surface structure. It focuses on an observation point at a suitable magnification during the observation stage, and the results were recorded using image processing techniques. Figure 9(a) depicts a micro-topography image of the sensing region's taper-in-taper structure.
Fig. 9. SEM images, (a) taper-in-taper fiber, (b) AuNP-, (c) AuNPs/MoS2-NPs-, (d) AuNPs/MoS2-NPs/CeO2-NPs-immobilized sensor structure.
In Fig. 9(b), many AuNPs are visible on the fiber cladding surface, and their distribution is uniform. Figure 9(c) depicts a uniform MoS2-NPs layer that covers the AuNPs-modified region. The results shown in Fig. 9(d) show that CeO2-NPs were successfully immobilized on the sensor surface. To further characterize the immobilization of the NPs samples, the EDS element compositions of the three sensor samples were analyzed separately. Due to the fact that an observation point is kept in SEM, the significant composition of elements within the observation area is analyzed. The SEM-EDS result for probe coated solely with AuNPs is shown in Fig. 10(a). According to the results, Si and O are the primary components of optical fiber, C is the primary component of organic substances such as MPTMS used in the coating process, and the presence of Au indicates the presence of AuNPs on the probe's surface. The detection of Mo, S and Au elements over AuNPs/MoS2-NPs-immobilized sensor structure probe confirmed the existence of MoS2-NPs along with AuNPs in Fig. 10(b). Similarly, the elements Ce, Au, Mo, and S were detected across the sensing region of AuNPs/MoS2-NPs/CeO2-NPs-immobilized sensor structure confirmed the presence of the NPs’ immobilization process in Fig. 10(c).
Fig. 10. SEM-EDS images, (a) AuNPs-, (b)AuNPs/MoS2-, (c) AuNPs/MoS2-NPs/CeO2-NPs- taper-in-taper fiber structure.
3.4 Measurement of analyte
A detection range of 10 U/L to 1000 U/L of ALT enzyme was chosen to ensure accurate detection of abnormal ALT levels in human serum. The target samples used in the experiment were concentrations of 0, 10, 50, 100, 250, 500, 750, and 1000 U/L. The ALT analyte can be determined after integrating the experimental device in accordance with Fig. 6. To begin, 5 mL of PBS solution was added at a temperature of 35°C to the ALT substrate tubes to dissolve the sample 1. After adding sample 1 to the reaction cell, the spectral result was recorded 20 minutes later, along with the peak wavelength of a blank sample. After removing the test liquid from the cell, the probe was rinsed with PBS and dried using nitrogen gas. Following that, sample 2 was prepared by adding 10 U/L of ALT enzyme solution to additional substrate tubes and reacting for three minutes at a temperature of 35 °C. Then, sample 2 was added to the measurement cell, and the LSPR spectrum was recorded using a computer. Finally, all samples were measured to obtain a set of LSPR spectra corresponding to the remaining ALT enzyme concentrations.
Figure 11(a) depicts the average of three sets of normalized wavelength spectrum from three different fiber probes for various ALT concentrations. Additionally, a zoomed version of ALT concentrations of 0 U/L, 100 U/L, and 1000 U/L is shown in the inset of Fig. 11(a) to clearly demonstrate the wavelength shift. The sensor's linearity response is depicted in Fig. 11(b). The plot has a linearity of 95.62%, indicating that as the ALT enzyme concentration increases from 10 to 1000 U/L, the shift of the resonant peak increases linearly. Additionally, the performance of the ALT sensor is primarily determined by its sensitivity and detection limit (LoD). The sensor's sensitivity, as determined by data analysis, is 4.1 pm/(U/L). The LoD values were determined by calculating the standard deviation (SD) of the stability test for the probe. As a result, the probe's limit of detection is 10.61 U/L. According to the data analysis, the proposed sensor is effective at diagnosing human liver injury.
3.5 Reproducibility and reusability test
This section discusses the reproducibility and reusability of the probe used to evaluate the fabrication technique for the ALT sensor. The reproducibility experiment is designed to ascertain whether probes constructed using the techniques described in the work perform similarly. Additionally, it is a critical metric for evaluating a technology's application potential. For this purpose, two distinct probes were used to measure the same concentration (500 U/L) at the same temperature. On the basis of the analysis of the two measured results, the LSPR spectrum shown in Fig. 12(a) was arranged. The result indicates that two groups of waveforms are highly coincident on the whole, and a similar conclusion can be drawn by magnifying them appropriately. The protocol for developing the probe is reproducible to a certain extent.
Reusability is a crucial standard for evaluating the cost, convenience, and efficiency of probes, as well as an important indicator of the sensor products’ clinical utility. In the reusability experiment, the ALT concentration in 50 U/L of sample solution was determined using a probe. After rinsing and drying, the measurement was repeated with the same samples. To facilitate comparison, the 250 U/L ALT samples were also quantified twice using the above-described method. Finally, two sets of data were collected, and the results are summarized in Fig. 12(b).
3.6 Stability and pH test
The stability test is used to determine the effect of multiple tests on the performance of the fiber sensor, that demonstrates the proposed sensor's potential application value. To conduct the measurement, a probe was tested ten times with 1×PBS samples and the corresponding LSPR spectra were recorded. The corresponding value of peak wavelength for each of the ten groups of data was analyzed and plotted as shown in Fig. 13(a). The result demonstrates that the peak wavelength values of ten-group data are nearly identical. Additionally, all measurements have a SD of 0.0145.
The pH test is a typical way for determining the optimal pH value for dissolving the enzyme by measuring samples in various pH solvents. Five groups of samples (pH4, acetic acid; pH6, ethanol; pH7.4, 1×PBS; pH8 and pH10, different concentration of potassium hydroxide) with varying pH values were prepared in this experiment. Each group received samples with two different concentrations ranging from 10 to 1000 U/L. The samples with low and high concentrations were measured sequentially and the corresponding LSPR spectra were recorded. All remaining samples were quantified using a similar method. After analyzing the data from the samples, the difference in the resonance wavelength peak for five groups was determined, and a correlation graph with the resolvent's pH value was drawn in Fig. 13(b). The wavelength peak shift is maximum at a pH of 7.4, indicating that 1×PBS is an appropriate solvent for the samples.
3.7 Selectivity test
The selectivity test is a critical component of evaluating the sensor's performance. Because clinical testing requires a sample of human blood, a variety of enzyme substances may interfere with the final results, adversely affecting their credibility. As a result, determining the specificity of the numerous enzyme substances found in blood is essential.
Thus, five distinct detection objects were prepared for this experiment: glycine, ascorbic acid, acetone, glucose, and ALT. Each group prepared two samples’ solutions using the highest (1000 U/L) and lowest (10 U/L) concentrations of the substance dissolved in warm-substrate tubes, respectively. The LSPR spectra of each group's samples were measured, and the waveforms were recorded after 20 minutes at the same room temperature. After measuring the low concentration sample solution, the probe was cleaned and dried, and then the high concentration sample solution was measured. Following the collection of measurement results of all samples, the difference in peak wavelengths for each group was calculated as shown in Fig. 14. As a result, we can conclude that the probe exhibits a high degree of specificity, as no obvious reaction occurs with the remaining four types of analytes. In other words, the probe binds to the ALT enzyme with a high degree of specificity.
3.8 Evaluation of sensing performance
To our knowledge, no fiber-based sensors have been developed, and most ALT sensors are based on spectroscopy, fluorescence, chromatography, or electrochemical techniques, all of them have significant drawbacks. For example, the spectrometric method has a low specificity for serum samples and inadequate anti-environmental interference capability. While fluorescent Alanine transaminase is frequently used, complicated equipment, a harsh environment, ionic strength, and a variety of other factors can all influence the results. Chromatography, like any other technique, has limitations because of its complicated operation and demanding test conditions. Electrochemical methods have a low signal resolution in comparison to the level of interference in serum samples. Additionally, Table 1 includes some representative sensor experimentation data. While the results demonstrate that these sensors have a good linear fit, some also have a linear range that is insufficient to cover the normal range of human enzyme activity. Furthermore, some tasks provide little information about the sensor's sensitivity or detection range. The taper-in-taper structure based on SMF is combined with the LSPR phenomenon in this work to create a biosensor that combines the unmatched advantages of biocompatibility, portability, economy, high sensitivity, reproducibility, and ease of operation. Without a doubt, the proposed LSPR sensor incorporating novel nanomaterials is an attractive and promising sensor that provides new alternatives to conventional ALT enzyme detection methods.
Table 1. Performance comparison of the Alanine aminotransferase (ALT) sensors.
4. Conclusion
In this experiment, a biosensor based on the LSPR principle was proposed for the detection of the ALT enzyme. To the author's knowledge, this is the first time, a taper-in-taper optical fiber structure has been developed for biosensing applications. The manufacturing process using the three-electrode semi-vacuum taper method and principles underlying the unique fiber structure are thoroughly discussed. Three distinct types of NPs solutions are characterized using absorption spectra and high-resolution HR-TEM images. Sensitivity and biocompatibility of the probe were increased by immobilizing the NPs in the sensing region. The probe's observations demonstrate the feasibility of the technology. To validate the effectiveness of the NPs coating protocols, the probe's micro-morphology was examined using SEM. Also, SEM-EDS was used to find the elemental composition of the sensing region. After that, an immobilized GluOx enzyme probe was used to detect the presence of ALT in the sample solution. Immobilization on the probe's surface is combined with the fact that ALT enzymes with varying activity concentrations can catalyze the substrate to produce varying concentrations of L-glutamate. By observing the shift in the resonant peak wavelength, the concentration of ALT enzyme activity was determined precisely. The results indicate a linear relationship between the concentration of ALT enzyme and peak wavelength of the resonance spectrum. The curve fitting values and limit of detection, as determined by the results, were 95.62% and 10.61 U/L, respectively. The sensitivity is 4.1 pm/(U/L) over a linear range of 10 to 1000 U/L. Additionally, the reproducibility, repeatability, stability, pH test, and selectivity of the novel structure-based sensor probe are test and found satisfactory. The results indicate that the developed sensor probe can detect the ALT enzyme and possesses some potential to diagnose the liver injury.
Funding
Double-Hundred Talent Plan of Shandong Province; Special Construction Project Fund for Shandong Province Taishan Mountain Scholars; Liaocheng University (31805180301, 31805180326, 318051901); Natural Science Foundation of Shandong Province (ZR2020QC061); Ministério da Educação e Ciência (UIDB/50025/2020, UIDP/50025/2020); Fundação para a Ciência e a Tecnologia (CEECIND/00034/2018).
Acknowledgments
This work was supported by Special Construction Project Fund for Shandong Province Taishan Mountain Scholars, China. C. Marques acknowledges Fundação para a Ciência e a Tecnologia (FCT) through the CEECIND/00034/2018 (iFish project), and this work was developed within the scope of the project i3N, UIDB/50025/2020 & UIDP/50025/2020, financed by national funds through the FCT/MEC.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. S. M. Lee, J. S. Park, Y. J. Han, W. Kim, and M. Y. Kim, “Elevated Alanine Aminotransferase in Early Pregnancy and Subsequent Development of Gestational Diabetes and Preeclampsia,” J Korean Med Sci 35(26), 198 (2020). [CrossRef]
2. V. W. Nielsen, J. G. Holm, A. Jrgensen, Y. Yao, H. C. Ring, and S. F. Thomsen, “Liver enzyme levels are associated with markers of systemic inflammation, disease severity, and metabolic syndrome in patients with hidradenitis suppurativa,” Health Sci Rep 4(3), 360 (2021). [CrossRef]
3. S. K. Kunutsor, T. A. Apekey, M. V. Hemelrijck, G. Calori, and G. Perseghin, “Gamma glutamyltransferase, alanine aminotransferase and risk of cancer: Systematic review and meta-analysis,” Int. J. Cancer 136(5), 1162–1170 (2015). [CrossRef]
4. Y. L. Du and K. S. Ryan, “Pyridoxal phosphate-dependent reactions in the biosynthesis of natural products,” Nat. Prod. Rep. 36(3), 430–457 (2019). [CrossRef]
5. E. D. Naser, T. Mohammad, S. Samira, and T. V. Sahar, “Review article: liver enzyme alteration,” Int. J. Med. Sci. 172(3), 367–379 (2012). [CrossRef]
6. S. Liangpunsakul and N. Chalasani, “Unexplained elevations in alanine aminotransferase in individuals with the metabolic syndrome: results from the third National Health and Nutrition Survey (NHANES III),” Am. J. Med. Sci. 329(3), 111–116 (2005). [CrossRef]
7. Z. Liu, S. Que, J. Xu, and P. Tao, “Alanine Aminotransferase-Old Biomarker and New Concept: A Review,” Int. J. Med. Sci. 11(9), 925–935 (2014). [CrossRef]
8. H. Chang-Jung, J. H. Wang, L. Dai, and C. C. Liu, “Development of an Electrochemical-Based Aspartate Aminotransferase Nanoparticle Ir-C Biosensor for Screening of Liver Diseases,” Biosensors 2(2), 234–244 (2012). [CrossRef]
9. A. A. Nanji, S. W. French, and J. B. Freeman, “Relationship between arterial oxygen tension and serum aminotransferases in morbidly obese patients with fatty liver,” Clin. Chem. 31(9), 1571–1572 (1985). [CrossRef]
10. H. L. Liu, M. A. Yong-Neng, P. Zhang, and Y. Zhang, “Diagnostic value of AFP,TBA,GGT,ALT joint detection on hepatocarcinoma,” MLS 2, 122–123 (2010). [CrossRef]
11. R. Kwan, P. Hon, and R. Renneberg, “Amperometric biosensor for rapid determination of alanine,” Anal. Chim. Acta 523(1), 81–88 (2004). [CrossRef]
12. R. I. Jirjis and M. David, “Comparison of spectrophotometric and electrophoretic measurements of glutamate dehydrogenase activity during mycelial growth and carpophore development in the basidiomycete Coprinus cinereus,” FEMS Microbiol. Lett. 6(2), 115–118 (1979). [CrossRef]
13. Kenneth E. Sherman, “Alanine aminotransferase in clinical practice. A review,” Arch. Intern. Med. 151(2), 260–265 (1991). [CrossRef]
14. M. Jamal, O. Worsfold, T. Mccormac, and E. Dempsey, “A stable and selective electrochemical biosensor for the liver enzyme alanine aminotransferase (ALT),” Biosens. Bioelectron. 24(9), 2926–2930 (2009). [CrossRef]
15. J. N. Sharma, D. K. Pattadar, B. P. Mainali, and F. P. Zamborini, “Size determination of metal nanoparticles based on electrochemically measured surface-area-to-volume ratios,” Anal. Chem. 90(15), 9308–9314 (2018). [CrossRef]
16. R. Shukla, V. Bansal, M. Chaudhary, A. Basu, and M. Sastry, “Biocompatibility of gold nanoparticles and their endocytotic fate inside the cellular compartment: a microscopic overview,” Langmuir 21(23), 10644–10654 (2005). [CrossRef]
17. Z. Bao and C. Q. Lan, “Advances in biosynthesis of noble metal nanoparticles mediated by photosynthetic organisms—A review,” Colloids Surf. B 184, 110519 (2019). [CrossRef]
18. J. K. Nayak and R. Jha, “Graphene-oxide coated Ag-island-based inline LSPR fiber sensor,” IEEE Photonics Technol. Lett. 30(19), 1667–1670 (2018). [CrossRef]
19. A. D. Chowdhury, F. Nasrin, R. Gangopadhyay, A. B. Ganganboina, K. Takemura, I. Kozaki, H. Honda, T. Hara, F. Abe, S. Park, T. Suzuki, and E. Y. Park, “Controlling distance, size and concentration of nanoconjugates for optimized LSPR based biosensors,” Biosens. Bioelectron. 170, 112657 (2020). [CrossRef]
20. T. Liu, H. Ding, C. Zhan, J. Huang, and S. Wang, “Simply and cost-effectively fabricated AuNP-based fusion spliced transmissive optical fiber LSPR probes,” Opt. Express 29(5), 7398–7409 (2021). [CrossRef]
21. S. Kaushik, U. K. Tiwari, S. S. Pal, and R. K. Sinha, “Rapid detection of Escherichia coli using fiber optic surface plasmon resonance immunosensor based on biofunctionalized Molybdenum disulfide (MoS2) nanosheets,” Biosens. Bioelectron. 126(1), 501–509 (2018). [CrossRef]
22. K. Li, W. Wu, Y. Jiang, Z. Wang, X. Liu, J. Li, D. Xia, X. Xu, J. Fan, and K. Lin, “Highly enhanced H2 evolution of MoO3/g-C3N4 hybrid composites based on a direct Z-scheme photocatalytic system,” Inorg. Chem. Front. 8(5), 1154–1165 (2021). [CrossRef]
23. A. Kaushik, P. R. Solanki, A. A. Ansari, S. Ahmad, and B. D. Malhotra, “A nanostructured cerium oxide film-based immunosensor for mycotoxin detection,” Nanotechnology 20(5), 055105 (2009). [CrossRef]
24. D. Patil, N. Q. Dung, H. Jung, S. Y. Ahn, M. J. Dong, and D. Kim, “Enzymatic glucose biosensor based on CeO2 nanorods synthesized by non-isothermal precipitation,” Biosensors and Bioelectronics 31(1), 176–181 (2012). [CrossRef]
25. S. Dong, N. Li, T. Huang, H. Tang, and J. Zheng, “Myoglobin immobilized on LaF3 doped CeO2 and ionic liquid composite film for nitrite biosensor,” Sensors and Actuators B: Chemical 173, 704–709 (2012). [CrossRef]
26. N. T. Nguyet, V. V. Thu, H. Lan, T. Trung, and P. D. Tam, “Simple Label-Free DNA Sensor Based on CeO2 Nanorods Decorated with Ppy Nanoparticles,” J. Electron 48(10), 6231–6239 (2019). [CrossRef]
27. H. Y. Lin, C. H. Huang, G. L. Cheng, N. K. Chen, and H. C. Chui, “Tapered optical fiber sensor based on localized surface plasmon resonance,” Opt. Express 20(19), 21693–21701 (2012). [CrossRef]
28. Y. Huang, Z. Tian, L.-P. Sun, D. Sun, J. Li, Y. Ran, and B.-O. Guan, “High-sensitivity DNA biosensor based on optical fiber taper interferometer coated with conjugated polymer tentacle,” Opt. Express 23(21), 26962–26968 (2015). [CrossRef]
29. A. Leung, P. M. Shankar, and R. Mutharasan, “A review of fiber-optic biosensors,” Sens. Actuators 125(2), 688–703 (2007). [CrossRef]
30. L. P. Sun, J. Li, Y. Tan, S. Gao, L. Jin, and B. O. Guan, “Bending effect on modal interference in a fiber taper and sensitivity enhancement for refractive index measurement,” Opt. Express 21(22), 26714–26720 (2013). [CrossRef]
31. Y. Tan, L. P. Sun, L. Jin, J. Li, and B. O. Guan, “Microfiber Mach-Zehnder interferometer based on long period grating for sensing applications,” Opt. Express 21(1), 154–164 (2013). [CrossRef]
32. L. Niu, C.-L. Zhao, H. Gong, Y. Li, and S. Jin, “Curvature sensor based on two cascading abrupt-tapers modal interferometer in single mode fiber,” Opt. Commun. 333(14), 11–15 (2014). [CrossRef]
33. T. Liyanage, M. Lai, and G. Slaughter, “Label-free tapered optical fiber plasmonic biosensor,” Anal. Chim. Acta 1169, 338629 (2021). [CrossRef]
34. J. Dey, B. Hazra, and M. Chandra, “Modulation of LSPR spectra and enhanced RI-sensitivity through symmetry breaking in hollow gold nanoprism,” J. Chem. Phys. 151(11), 114706 (2019). [CrossRef]
35. J. Zhao, D. Huo, X. Geng, J. Bao, J. Hou, Z. Shui, H. Yang, Y. Qi, Y. Hu, M. Yang, and C. Hou, “3D MoS2-AuNPs carbon paper probe for ultrasensitive detection and discrimination of p53 gene,” Biosens. Bioelectron. 332(1), 129480 (2018). [CrossRef]
36. M. Qiu, R. Zhang, Y. C. Li, and Z. S. Zhou, “Effects of CeO2 Addition on Tribological Performance of MoS2 Based Composite Coatings,” China Surface Engineering 30(2), 106–112 (2017).
37. T. S. Song, L. Fu, N. Wan, J. Wu, and J. Xie, “Hydrothermal synthesis of MoS2 nanoflowers for an efficient microbial electrosynthesis of acetate from CO2,” J. CO2 UTIL 41, 101231 (2020). [CrossRef]
38. M. Li, R. Singh, C. Marques, B. Zhang, and S. Kumar, “2D Materials assisted SMF-MCF-MMF-SMF based LSPR Sensor for Creatinine Detection,” Opt. Express 29(23), 38150–38167 (2021). [CrossRef]
39. S. Kumar, B. K. Kaushik, R. Singh, N. K. Chen, and B. Zhang, “LSPR-based cholesterol biosensor using a tapered optical fiber structure,” Biomed. Opt. Express 10(5), 2150–2160 (2019). [CrossRef]
40. S. M. Ansar, S. Chakraborty, and C. L. Kitchens, “pH-Responsive Mercaptoundecanoic Acid Functionalized Gold Nanoparticles and Applications in Catalysis,” Nanomaterials 8(5), 339 (2018). [CrossRef]
41. F. Qu, H. Lu, M. Yang, and C. Deng, “Electrochemical immunosensor based on electron transfer mediated by graphene oxide initiated silver enhancement,” Biosens. Bioelectron. 26(12), 4810–4814 (2011). [CrossRef]
42. J. S. Pieper, T. Hafmans, J. H. Veerkamp, and T. Kuppevelt, “Development of tailor-made collagen-glycosaminoglycan matrices: EDC/NHS crosslinking, and ultrastructural aspects,” Biomaterials 21(6), 581–593 (2000). [CrossRef]
43. C. Y. Si, Z. Z. Ye, Y. X. Wang, L. Gai, J. P. Wang, and Y. B. Ying, “Rapid detection of Escherichia coli O157 : H7 using surface plasmon resonance (SPR) biosensor,” SPECTROSC SPECT ANAL 31(10), 2598–2601 (2011). [CrossRef]
44. S. K. Arya, A. Dey, and S. Bhansali, “Polyaniline protected gold nanoparticles based mediator and label free electrochemical cortisol biosensor,” Biosens. Bioelectron. 28(1), 166–173 (2011). [CrossRef]
45. G. Zhu, L. Singh, Y. Wang, R. Singh, B. Zhang, F.-Z. Liu, B. K. Kaushik, and S. Kumar, “Tapered optical fiber-based lspr biosensor for ascorbic acid detection,” Photonic Sens. 11(4), 418–434 (2021). [CrossRef]
46. A. N. Ellina, R. Ab, A. Jaafar, Y. N. Azah, and S. Hamidah, “Fluorescence biosensor based on encapsulated quantum dots/enzymes/sol-gel for non-invasive detection of uric acid,” J. Lumin. 202, 309–315 (2018). [CrossRef]
47. J. M. Fernández-Romero, M. D Luque de Castro, and M. Valcárcel, “Determination of alanine aminotransferase in human serum in an open-closed flow injection configuration,” J. Biotechnol. 14(1), 43–52 (1990). [CrossRef]
48. G. S. Xuan, W. O. Sang, and E. Y. Choi, “Development of an electrochemical immunosensor for alanine aminotransferase,” Biosens. Bioelectron. 19(4), 365–371 (2003). [CrossRef]
49. T. Sugaya, S. Naito, S. Yonezawa, F. Morishita, and T. Kojima, “Simultaneous assay of aspartate aminotransferase and alanine aminotransferase by flow injection method using immobilized enzyme reactors,” Anal. Sci. 4(6), 579–582 (1988). [CrossRef]
50. M. J. Song, D. H. Yun, and S. I. Hong, “An electrochemical biosensor array for rapid detection of alanine aminotransferase and aspartate aminotransferase,” Bioscience, Biotechnology, and Biochemistry 73(3), 474–478 (2009). [CrossRef]
51. S. Ohmori, Y. Tanaka, M. Ikeda, and K. Hirota, “A sensitive determination of α-keto acids by gas-liquid chromatography and its application to the assay of l-glutamate dehydrogenase and aminotransferases,” Anal. Biochem. 112(2), 204–212 (1981). [CrossRef]
