1. Introduction

Histamine is a biological amine produced due to the decomposition of histidine by decarboxylase [1] is a potentially harmful nitrogen-containing low molecular weight compound, that extensively exists in animals and plants, especially in aquatic products like fish. Since the population will expand substantially by 2050, there is a need for high-quality protein to feed humans, making aquaculture more critical than ever [2]. However, fish meat is particularly critical and should not be stored for an extended period of time. Under the influence of external histamine-producing bacteria and endogenous histidine decarboxylase, histidine leads to accumulation of a huge amount of histamine if the fish is not handled promptly and properly [3]. Therefore, the production of histamine is closely related to the spoilage process of fish. This highly toxic chemical binds to the histamine receptor associated with G protein, result in a variety of allergic inflammatory response diseases, like headaches, nausea, variations in blood pressure, respiratory disorders, and in rare, even death [4–7]. Higher histamine concentration in the human body i.e., more than 9 mM, also induces histamine-induced food poisoning [8]. The extensive therapeutic signs of histamine poisoning are mental malaise, allergies, flushing of the skin, vomiting, and diarrhea [9,10]. Consequently, it is essential to formulate guidelines and rigorously control the amount of histamine in seafood. The Food and Drug Administration (FDA) emphasized that the acceptable value of the chemical index for fish contamination is 450 µM and that the recommended value of the threat threshold is between 1.8 and 4.5 mM [8,10]. Histamine has become an important indicator for determining the safety and freshness of fish meat during food production, storage, and transportation. Therefore, it is crucial to develop an accurate, simple, and effective method for histamine detection in diverse ways including directly in water fish tanks and also during storage and transportation till market.

In recent years, researchers have shown great research interest in optical fiber-based sensors due to its distinctive advantages, such as light weight, high sensitivity, high interference immunity. Esposito [11] discussed the fiber optic chemical sensors based on long period grating (LPG) technology. It covered the design, fabrication and characterization of these sensors, as well as coating materials and applications. It aimed to provide a comprehensive analysis of current state-of-the-art and future research directions. In another work, Esposito et al. [12] proposed an optical biosensor based on fiber optic LPG for label-free detection of vitamin D. The biosensor used an LPG in a double cladding fiber coated with GO and a specific antibody for detection of vitamin D. This sensor can detect vitamin D with a highly sensitive, selective and stable method. Similarly, a new method [13] for improving the sensitivity of long-period gratings (LPGs) to surrounding refractive index (SRI) using double-clad fibers with a W-type refractive index (RI) distribution is reported. The authors show that by reducing the fiber diameter, they can induce a mode transition phenomenon that improves SRI sensitivity by a factor of 40 to 60 compared to an unetched fiber. They demonstrate this experimentally with LPG written in the pure quartz cores using fluorine-doped inner cladding and silica outer cladding fibers, and focus on biological applications of SRI sensing in aqueous-like environments.

There are many techniques existing for histamine determination [14–18]. But traditional histamine detection methods have many limitations, such as time-consuming operations, high cost, complex operations, consequently, these methods can not satisfy the practical demands. However, localized surface plasmon resonance (LSPR)-based fiber optic biosensors provide numerous advantages over conventional detection methods, such as small size, high sensitivity, and real-time detection [19,20]. Due to the shape and size-dependent properties of nanoparticles, its resonance feature can be tailored from visible to near infrared region. Further, fiber-based nanoparticle system can be used in wavelength interrogation as well as in intensity interrogation by controlling the surface coverage of nanoparticles around the surface of the optical fiber. These advantages make LSPR-based optical fiber biosensors extremely promising for application in numerous sectors, such as biochemical detection, food safety, illness diagnosis, and environmental monitoring. Li et al. [21] proposed an LSPR-based optical fiber biosensor for detecting creatinine in aquaculture. At present, some researchers have developed LSPR-based fiber-optic sensor to detect the content of p-cresol in water [22,23].

Due to its immense promise in numerous detecting sectors, optical fiber biosensors based on the LSPR effect induced by precious metal nanoparticles (MNPs) have sparked a plasma study and application with the advent of nanotechnology. The remarkable sensitivity of MNPs induced LSPR to ambient RI changes enables this form of biosensor possible. It is known that when light enters an optical fiber at a specific angle, total internal reflection (TIR) will occur during transmission. At the interface of two separate media, the waves will be formed that return to the optically dense media after traveling the wavelength order distance in the optically sparse media. The amplitude of this wave decreases exponentially along the vertical direction of the interface, and it is known as an evanescent wave (EW). The produced EWs will excite the collective oscillation of valence electrons on the surface of the metal, so forming surface plasma waves. The evanescent field formed at the interface between the fiber and metallic nanoparticles interacts with the MNPs, that excites the surface plasmons within the MNPs, leading to the generation of a high surface field. LSPR sensors use the change in peak wavelength in the absorption spectra of MNPs as the sensing mechanism. In addition, the probe is decorated with the appropriate ligand and the target analyte is coupled with the ligand on the nanoparticles, resulting in a change in the RI around the nanoparticles. This would result in a shift in the LSPR resonance peak position in the extinction spectrum [24]. Thus, the small change in local surface RI induced by biomolecular reaction is transformed into a wavelength shift of the observable LSPR peak absorption wavelength, and biochemical detection is achieved via spectrum analysis.

Due to their unique properties, MNPs and carbon nanomaterials of varying diameters have garnered considerable interest from scientists in recent years. AuNPs have the properties of easy synthesis, controllable morphology and particle size, more intensive extranuclear electrons, and better biocompatibility, and are considered to be the best choice for immobilization materials [25]. Therefore, AuNPs have been widely used in the development of biosensors. Similarly, graphene oxide (GO) has been expansively used as a matrix support material for biosensors due to its superior structural specificity, including large specific surface area, strong adsorption capacity, unique electron mobility and plasmon effect. Nowadays, multi-walled carbon nanotubes (MWCNTs) are ideal building blocks for molecular systems, with excellent planar conductivity, high specific surface area, and surface energy [26]. Compared with one-dimensional carbon nanotubes, GO, a honeycomb lattice structure formed by tightly packed carbon atoms, is an excellent two-dimensional (2D) reinforcing material. As the carrier of layered materials, GO layer is used in polymers. The surface of GO layer contains epoxy groups, and there are a significant number of hydroxyl and carboxyl groups at the edge of the sheet, that has good dispersibility and hydrophilic [27]. Combining the one-dimensional tubular structure of MWCNTs with the 2D layered structure of GO will improve the conductivity of the composite, and also obtain higher sensitivity and shorter response recovery time.

In this work, a novel humanoid-shaped tapered optical fiber (HTOF) structure-based LSPR sensor probe is fabricated and developed using single-mode fiber (SMF) for the detection of histamine. The sensor functionalized with GO/MWCNTs effectively increased the specific surface area of the fiber probe, providing more attachment sites for AuNPs immobilization and diamine oxidase (DAO) enzyme functionalization. According to the experimental results, the linear range, sensitivity, and limit of detection (LoD) are 0-1000 µM, 5.5 nm/mM, and 59.45 µM, respectively. Based on the LSPR phenomenon, this sensor realizes a high sensitivity, low LoD, fast response optical fiber biosensor for histamine detection.

2. Materials and methods

2.1 Materials

The SMF (8.2 µm, 125 µm) purchased from EB-link Technologies Co, Shenzhen was used to fabricate the HTOF structure. Hydrogen tetrachloroaurate (HAuCl₄), trisodium citrate and deionized (DI) water were used to synthesis the AuNPs solution. GO was synthesized from graphite powder, potassium permanganate (KMnO₄), sulfuric acid (H₂SO₄), sodium nitrate (NaNO₃) and hydrochloric acid (HCl) using modified Hummer’s method. For the cleaning of HTOF surface and the immobilization of nanomaterials (AuNPs, GO/MWCNTs), acetone, hydrogen peroxide solution (H₂O₂, 30%), concentrated sulfuric acid (H₂SO₄, 98%), (3-mercaptopropyl) trimethoxysilane (MPTMS) and ethanol were utilized. To enzyme functionalize over the nanomaterials-immobilized probe, 11 mercaptoundecanoic acid (MUA), N-hydroxysuccinimide (NHS), N-(3-dimethylaminopropyl) - N-ethanediimine hydrochloride (EDC) and phosphate buffered saline (1×PBS, pH = 7.4) were required. N,N-dimethylformamide (DMF) solution was used for preparing the GO/MWCNTs solution. MWCNT was purchased from Xianfeng Nano, Nanjing. The diamine oxidase (DAO) enzyme was purchased from Sigma-Aldrich, Shanghai and used to add the selectivity property on probe. Several biomolecules, such as histamine, tryptamine, spermine, tyramine and putrescine were purchased from Macklin, Shanghai for sensing experiment.

2.2 Instrument and measurement

The following high-precision instruments were used for the development of the sensing probe. The HTOF structure was fabricated using the 3SAE combiner manufacturing system (CMS, USA). A UV-Visible spectrophotometer was used to evaluate the absorption spectrum of AuNPs solution in order to estimate the absorbance wavelength. Scanning electron microscopy (SEM, Carl Zeiss Microscopy, Germany) scanned the samples by emitting a high-energy electron beam to observe the nanomaterials coating on the surface of the sensing probe. High-resolution transmission electron microscopy (HR-TEM, Talos L120C, Thermo Fisher Scientific, USA) was used to confirm the microscopic distribution of nanomaterials in the synthesized solution. In addition, the experimental set-up consist a tungsten-halogen light source (HL-2000, Ocean Optics, USA) that emits a broad spectrum, and a spectrometer (USB2000+, Ocean Optics, USA) was used to collect experimental data to study the optical transmission characteristics inside HTOF-based sensor probe.

2.3 Sensing mechanism of the probe

For LSPR-based biosensors, the realization of excellent sensing performance is inseparable from the formation of an effective evanescent field. The evanescent field is conducive to stimulating the LSPR effect, resulting in RI-related absorption peaks, that in turn enable analyte sensing. However, it is difficult to form a strong evanescent field with conventional fiber system that largely decides the performance of the sensing system mediated by evanescent field. The common method to enhance evanescent field is to change the fiber structure, that can be roughly divided into the following three types. One is to change the transmission path of light in the optical fiber by bending/polishing the optical fiber structure, such as U-shaped and D-shaped optical fiber structures [28,29]. Although these sensing probes are simpler to implement, their repeatability and sensitivity are limited. The second is that the core mode mismatch is caused by the fusion of fibers with different core diameters, so that more light enters the cladding to improve the evanescent field, for example, SMF-multi-mode fiber (MMF)-SMF [30] and other different core-mismach/core-offset structures. The third is to enhance the evanescent field by changing the diameter of the fiber and reducing the thickness of the core/cladding, such as tapered fiber and etched fiber [31].

This work employs the tapered fiber due to its low cost, high sensitivity, compact size, and relatively simple production. Moreover, it can reach the cladding interface to interact with surrounding targets, and it is very sensitive to RI changes of analytes in surrounding media, consequently modifying the transmission mode, spectrum, light intensity, and other detectable features of the output signal. In comparison to the conventional single-taper fiber structure, the HTOF structure added the taper treatment to the single-taper, thereby widening the range of the sensing region that can generate stronger evanescent field and providing greater immobilization space for nanomaterials and enzyme functionalization. Figure 1 shows the schematic of HTOF sensor structure.

 

Fig. 1. Schematic of humanoid tapered optical fiber (HTOF)-based sensor probe.

Download Full Size | PDF

During the taper fabrication process, the size ratio of the core and the cladding remains basically unchanged. The SMF (8.2/125 µm) is processed to a taper waist diameter of 40 µm and a core diameter of 2.6 µm [32]. At this time, the original core will no longer be able to confine light waves, the entire taper region can be defined as the new core, and the external medium acts as the cladding [33]. Here, the nanomaterials and specific enzymes are immobilized on the surface of the probe by covalent binding. When light passes through the sensing region, the nanomaterials absorb light to produce the LSPR effect, and show the resonance peak on the detector. When the enzyme functionalized on the nanomaterials specifically binds to the analyte to be measured, the RI near the probe changes, resulting in a peak resonance wavelength shift. That is related to the analyte concentration to be measured in order to achieve quantitative detection of the analytes to be measured. When the effective RI of an external medium change, the resonance wavelength shift can be expressed as [34];

2.4 Fabrication of sensor probe

The HTOF sensor probe used in this work was fabricated from the standard SMF, and the instrument used for tapering is CMS with advanced processing performance. The thermal repeatability of thermally stable plasma in CMS is greater than 10 times that of any existing arc technology. The unique heating method produces a highly adjustable plasma field, completely surrounds the optical fiber to achieve a highly uniform heat distribution, and is capable of achieving a highly symmetric ultra-low loss tapering [35]. CMS supports a unique three-electrode operation mode to maximize the repeatability of the heating area. The whole process of CMS taper drawing is controlled by the debugged computer program, and different taper structures correspond to different programs. There are many important parameters in the program, mainly including initial power, waist power, taper diameter, operating speed and vacuum value, that are determined through a multiple test until a perfect tapered structure is drawn. The preparatory work [36] before taper drawing included the pretreatment of optical fiber and the calibration of CMS working platform. The calibration process of CMS working platform included three parts, namely, the exclusive automatic alignment of pitch, pitch and yaw, with a resolution of less than 0.01 °. After calibration, placed the pretreated optical fiber on the working platform and fasten the clamps at both ends to ensure that the tensions at both ends of the optical fiber are the same. The internal structure of CMS and the three-electrode heating process are shown in Fig. 2 (a) and (b). The fabrication of the HTOF structure is mainly controlled by two programs, including program 1 (initial diameter 125 µm, waist diameter is 40 µm) and program 2 (initial diameter 80 µm, waist diameter 60 µm). The tapering process from the initial conventional SMF optical fiber to the HTOF structure is shown in Fig. 2(c).

 

Fig. 2. (a) Internal structure of CMS instrument, (b) three electrode heating process, and (c) HTOF structure fabrication process.

Download Full Size | PDF

2.5 Synthesis process of AuNPs and GO/MWCNTs

The Turkevich method [37] was used to synthesize the AuNPs with a particle size of 10 nm, and the modified Hummer’s method [38] was utilized to synthesis the GO. The method for preparing GO/MWCNTs composites is divided into the following steps. First, 0.0375 g of GO and 0.0125 g of pretreated MWCNTs were weighed, respectively, and 100 mL of DI water was added to a glass bottle to obtain a mixed solution of these two nanomaterials. Then, 1 mL of hydrazine solution (35 wt%) and 7 mL of ammonia solution (28 wt%) were added to the mixed solution, respectively. Thereafter, heated the above solution at a temperature of 80°C and continue heating at 300 rpm and stirring for 30 min. Subsequently, the mixed solution was treated in a 90°C water bath for 1 hour, and then centrifuged at 3500 rpm (to completely disperse the particles in the solution) for 1 hour. Finally, the GO/MWCNTs powder was obtained by filtration and drying, that was ground and dissolved in DMF to obtain a fresh 0.5 mg/mL of GO/MWCNTs hydride solution.

2.6 Nanomaterials and enzyme functionalization

NPs Immobilization and enzyme functionalization by amino coupling are key factors to determine the performance of optical fiber biosensors. Therefore, it is necessary to develop a stable and specific sensing layer. The entire immobilization process is divided into five steps. Before the nanomaterials were immobilized on the probe surface, the fiber surface should be cleaned thoroughly. Step 1: to remove any remaining particles from the surface of the optical fiber's tapering portion, the optical fiber was washed in an acetone solution for 20 minutes, followed by a thorough cleaning with DI water and drying in a nitrogen gas environment. Step 2: treatment of the probe with a mixed solution of sulfuric acid and 30% hydrogen peroxide solution (piranha solution, with a volume ratio of 7:3) for 30 min. Step 3: probes were rinsed with DI water and dried at 60°C for 30 min. After treatment, the silanol groups on the tapered fiber surface were fully exposed, that facilitated the adhesion of the silane reagent. Thereafter, fiber was immersed in the freshly prepared GO/MWCNTs mixed solution for 10 min. Then, fiber was dried at 80°C for 30 min. Repeated this step 3 times, and after 3 times of soaking and drying, the probe surface consists a uniform coating by tightly wrapping on the surface of the uncoated tapered area by Van-der Waals force. Step 4: the probes were placed in the 1% MPTMS ethanol solution for 12 hours, the fiber was rinsed with ethanol and dried with nitrogen gas. MPTMS as a coupling agent, can modify a uniform thiol monolayer on the surface of the probe. MPTMS functional methoxy group (CH₃O-) can undergo alcoholysis reaction with silyl hydroxyl group through hydrolysis, thereby forming thiol-terminated self-assembled monolayer on the surface of tapered fiber. Step 5: fully immersed the probe in the synthesized AuNPs solution for 48 h, followed by rinsing the probe with ethanol solution and dry with nitrogen gas. One end of the coupling agent MPTMS reacted with the AuNPs to form a chemical bond, and Au easily interacted with the sulfhydryl group (-SH) to form a stable Au-S covalent bond, that promoted the firm adhesion of AuNPs.

After completing the above nanomaterial immobilization process, the enzyme functionalization process was divided into three steps. Step I: the probe was immersed in MUA ethanol solution (10 mL, 0.5 mM) to do the carboxylation process for 5 hours. Step II: immerse the probe in a mixture of 5 mL of EDC (200 mM) and NHS (50 mM) for 30 min to fully activate the carboxyl group. EDC converts the carboxyl group into an amino-active intermediate that can be condensed with the amino group in the enzyme. However, the amino-active intermediate is unstable and will undergo hydrolysis to reconvert the carboxyl group. By combining EDC and NHS, EDC and NHS will convert the carboxyl group into an amino-active NHS ester, that will not hydrolyze and can condense with the amino group in the enzyme compared to the amino-active intermediate, thus increasing the efficiency of functionalized diamine oxidase on the HTOF surface. Step III: the probe was immersed in a freshly prepared DAO enzyme (160 mg/mL) solution for 6 hours, and the amino group of the DAO and the activated carboxyl group formed a covalent bond to realize the functionalization of the enzyme. Figure 3 depicts the nanomaterial immobilization and enzyme functionalization of the HTOF structure.

 

Fig. 3. AuNPs/GO/MWCNTs-immobilization and DAO enzyme functionalization over HTOF sensor structure.

Download Full Size | PDF

2.7 Preparation of histamine solution

In order to prepare 1000 µM histamine stock solution, dissolved 5.56 mg histamine powder in 50 mL of 1×PBS buffer (pH = 7.4). Other concentrations of histamine were diluted with PBS (pH = 7.4) solution to prepare a histamine concentration of 0 µM, 50 µM,100 µM, 400 µM, 600 µM, 800 µM. The histamine concentration should be measured from low concentration to high concentration to minimize the experimental error caused by residual solution.

2.8 Experimental setup

The optical signal was generated with a tungsten-halogen light source (HL-2000), and the output spectrum was received through a spectrometer in a spectral range of 200-1000 nm. When the light passed through the sensing area, the NPs absorb the light and produced a resonance phenomena. So when specific enzyme immobilized on the nanomaterials reacts with histamine, the RI around the AuNPs on the sensing area changes, resulting in a shift in the LSPR resonance peak. During the detection procedure, the prepared histamine solution was injected into the reaction holder in order of concentration from low to high, and the data corresponding to each concentration was recorded once the spectrum output was stable. To reduce errors, the sensor probe must be washed with PBS solution before testing new samples. Figure 4 depicted the chemical reaction mechanism of histamine as well as effective optical measuring equipment for detecting different concentration of histamine solutions.

 

Fig. 4. Histamine reaction process and experimental setup for detection of histamine solution using developed sensor.

Download Full Size | PDF

3. Results and discussions

3.1 Optimization of sensor probes

The key to CMS taper drawing technology was the adjustment of the taper program parameters (electrode power, taper waist diameter, motor moving speed, etc.). Through a large number of optimization experiments, the optimal parameters of the taper have been adjusted, and the tapered fiber with high precision and good repeatability can be processed. The diameter of the tapered region has a major impact on the characteristics of the tapered fiber structure, and the size of the tapered fiber affects its sensitivity. Although the reduced diameter of the sensing area has an improved sensitivity, too thin tapered waists are prone to breakage during probe functionalization or measurement. Therefore, considering the feasibility of the experiment, tapered optical fiber with diameters of 40 µm and 60 µm were finally chosen to develop the sensor probe. Here, Fig. 5(a) was a high magnification SEM image of our sensor probe. Figure 5(b) showed the scanning results of three different HTOF diameters, it can be seen that the diameters of the tapered waist area were uniform. This result validated the high repeatability of the fabrication of HTOF sensor structures. This structure is also shaped like a human, hence the name “Humanoid Shaped Optical Fiber”. Figures 5(c) and (d) show SEM scanning images of HTOF morphology with four tapered positions labeled.

 

Fig. 5. (a) SEM image of NPs-immobilized -, (b) diameter analysis of -, (c) and (d) SEM images of - HTOF structure.

Download Full Size | PDF

3.2 Characterization of NPs

Since the size and shape of the AuNPs are critical to the sensing results, the synthesized AuNPs solution was characterized by UV-Visible spectrophotometer, and the absorption spectrum was shown in Fig. 6(a). The absorption peak appears at 519 nm, indicating that AuNPs with a diameter of about 10 nm have been synthesized. Figure 6(b) is the result of observing the morphology of AuNPs in solution with HR-TEM. It can be seen that the synthesized NPs were spherical and uniform in size, that is more effective for the excitation of LSPR phenomenon. The histogram was shown in Fig. 6(c), the results showed that average size of synthesized AuNPs were 10.5 ± 0.5 nm. Figure 6(d) depicted that a large number of GO/MWCNTs nanomaterials were found in our synthesized nanomaterial solution, that was advantageous for expanding the enzyme's binding site.

 

Fig. 6. (a) Absorbance spectrum -, (b) TEM image-, (c) histogram of - AuNPs and (d) TEM image of GO/MWCNTs.

Download Full Size | PDF

3.3 Characterization of NPs-immobilized structure

In order to verify the immobilization of NPs on HTOF structure, the probe was characterized by SEM. SEM results clearly showed the immobilization of AuNPs/GO/MWCNTs on the probe surface, as shown in Fig. 7(a). The probe was analyzed by energy spectrum to prove the existence of Au. It can be seen from the Fig. 7(b) result that it contained a large number of Au and C elements. This also confirmed that AuNPs and GO/MWCNTs were effectively immobilized on the probe surface. Other elements, Si and O are present due to the composition of silica ($Si{O_2}$) optical fiber.

 

Fig. 7. (a) SEM image and (b) EDX of AuNPs/ GO/MWCNTs-immobilized optical fiber sensor probe.

Download Full Size | PDF

3.4 Measurement of analytes

In this section, the optical fiber probes were utilized to measure the different concentrations of histamine analyte. To reduce experimental error, three sensors were employed to measure each concentration of histamine solution. The probe sensing area was rinsed with PBS solution for 10 minutes and dried naturally. Then, appropriate amount of histamine solution was taken with a concentration of 10 µM and added it to the reaction holder. After stabilizing the LSPR spectrum and recording the data, the probe was rinsed with PBS solution and dried. The other histamine solutions were measured one at a time, from low to high concentrations, and the LSPR spectra for each concentration were recorded. The above measurement was repeated with three different probes, and the transmitted intensity spectrum was normalized after averaging the LSPR spectrum, as shown in Fig. 8(a). It can be concluded that as the histamine concentration increases, the peak wavelength shifted to a longer wavelength. The appearance of by-products near the probe sensing area shifted the LSPR spectra due to the catalytic effect of DAO enzyme with histamine in the reaction holder [39]. The increase of RI eventually led to the red shift of the spectrum. The chemical reaction of the substance to be measured under the catalysis of DAO enzyme is shown in Fig. 4. The linear curve fitted by the resonance wavelength is shown in Fig. 8(b). The probe's linear fitting coefficient is 0.99 in the 0-1000 µM range, indicating that the sensor has good linearity. The approximate linear change of peak wavelength with histamine solution can be expressed as:

 

Fig. 8. (a) LSPR sensing spectrum, (b) linearity plot of HTOF sensor probe.

Download Full Size | PDF

Sensitivity is the slope of the linear fitted curve, and standard deviation (SD) is measured ten-times using the same sensor probe for PBS solution, and then the SD was calculated with the peak wavelength of all measurements. In practical biosensor detection, the LoD is an important index to assessing the performance of biosensors. It is defined as the minimum detectable signal caused by the detected analyte interaction, and the formula is expressed as:

After calculation, the LoD of the developed histamine sensor is 59.45 µM.

3.5 Stability and pH test

The stability test is to investigate whether the initial wavelength position of the sensor probe is stable during multiple measurements. In the application of biosensors that require multiple continuous monitoring, stability is one of the important characteristics that cannot be ignored. First, PBS solution was added, followed by recording of LSPR spectrum, further probe surface was dried, and then PBS solution was added again. PBS solution was tested ten-times continuously with the same probe. The results were analyzed according to the peak resonance wavelength. The stability test results are shown in Fig. 9(a). According to the analysis of the results based on the peak resonance wavelength, the resonance wavelength of the probe was almost the same, and the SD was 0.109, indicating that the probe is stable and reliable.

 

Fig. 9. (a) Stability, and (b) pH test of developed sensor.

Download Full Size | PDF

Similarly, the purpose of pH testing was to observe the solubility of histamine in different pH environments. With the help of acetic acid and KOH solutions, different solvent environments of pH = 3, pH = 6, pH = 10 and pH = 13 were prepared, respectively. And in the above different solvent environments, 0 µM and 1000 µM histamine solutions were prepared, respectively. The test was carried out in the order of pH values from low to high. After the test, fully cleaned the surface of the sensor probe with a blank solution of the same pH value and dried it naturally. The same measurement method was performed for the solutions at other pH, and the results were shown in Fig. 9(b). The abscissa represented the liquid environment of different pH, and the ordinate was the resonance wavelength shifts between the corresponding concentrations at each pH. The results show that the maximum drift occurs in a liquid environment with a pH of 7.4. Therefore, PBS (pH = 7.4) was selected as the best solvent for the experiment and it is in line with pH of water fish tanks (pH around 7.3 and 7.5) to measure directly.

3.6 Reproducibility and reusability test

Reproducibility refers to multiple measurements of histamine solution of the same concentration using different sensor probes. The histamine solution with a concentration of 800 µM was tested with three same probes. The data was recorded after the LSPR spectrum was stabilized. Figure 10(a) showed the reproducibility test results of the probes. For the same concentration of histamine, different probes have the same resonance wavelength, indicating that the probes have high reproducibility.

 

Fig. 10. (a) Reproducibility, and (b) reusability results of developed sensor.

Download Full Size | PDF

Reusability is another important indicator of the sensor probe. If the probe can be used repeatedly, it can reduce the sensing cost in application. The reusability was measured twice with the same sensor probe at concentrations of 400 µM and 1000 µM histamine in order to examine the probe’s performance in repeated use. First, tested the histamine with 400 µM concentration, and recorded the data after the LSPR spectrum stabilized. After recording, rinsed the probe surface thoroughly with PBS, dried the probe and then measured the 400 µM concentration histamine again. Similarly, 1000 µM concentration histamine solution was measured twice in a similar manner and results were plotted in Fig. 10(b). The appearance of the same peak resonant wavelength for the same concentration indicated that the developed sensor probe can be reuse in essence, confirming the good reusability of the sensor.

3.7 Selectivity test

One of the most important factors for sensing performance is the sensor's selectivity. The purpose of the selectivity test is to confirm the sensor's ability to recognize histamine specifically in the presence of complex biological interference substances. Five interfering substances including tryptamine, spermine, tyramine, putrescine, and L-α-alanine were selected for selectivity testing in the linear range (0-1000 µM). For this purpose, detected the lowest concentration (0 µM) and highest concentration (1000 µM) solution of the above five interfering substances, and measured the peak resonance wavelength and calculated. The selectivity test results (as shown in Fig. 11) showed that in the presence of histamine the peak wavelength shifts of the LSPR spectrum reached its maximum, whereas other biomolecules did not induce any significant change in LSPR spectrum. This is due to DAO enzyme's specific catalytic effect on histamine (as shown in Fig. 4), that resulted in significant changes in RI. Therefore, it was proved that the sensor probe has high selectivity for histamine.

 

Fig. 11. Selectivity test of the DAO/AuNPs/GO/MWCNTs-immobilized HTOF sensor.

Download Full Size | PDF

3.8 Evaluation of sensing performance

Till now, some existing techniques for detecting histamine, such as fluorescence method, electrochemical method, surface-enhanced Raman scattering (SERS) and colorimetric method have been developed, but these techniques have drawbacks such as time-consuming, tedious operation steps, and high material cost. Relatively speaking, the advantages of the sensor developed in this experiment are its simplicity, low cost of production, and broad linear range. Table 1 was a comparison of sensor performance, including the use of nanomaterials, mechanisms, linear range, sensitivity and LoD. The developed LSPR sensor combined with novel nanomaterials has great sensitivity and low fabrication cost, that paves the way for a new histamine detection strategy.

Table 1. Comparison of the proposed sensor’s performance to that of existing sensors

View Table

4. Conclusion

In this work, a novel HTOF fiber-optic biosensor based on the LSPR effect is developed for the detection of histamine solutions with varying concentrations. The tapered fiber penetrates the intrinsic TIR, allowing it to obtain sufficient evanescent field to stimulate the LSPR effect. Utilizing AuNPs and the nanomaterials GO/MWCNTs enhanced the sensitivity of the probe, while the DAO enzyme improved its specificity. The experimental results showed that the histamine content was linearly proportional to the resonant peak wavelength in the 0-1000 µM range. The linear fitting coefficient of the probe is 0.99, the sensitivity is 5.5 nm/mM, and the LoD is 59.45 M. Furthermore, the repeatability, stability, pH test, and selectivity of the sensing probes all are assessed and found to be adequate. As a result, the novel sensor probe proposed in this work has significant application potential for detecting histamine in the field of food safety.