A chemiluminescence sensor for determination of lysozyme using magnetic graphene oxide multi-walled carbon nanotube surface molecularly imprinted polymers

Abstract

In this paper, a new chemiluminescence (CL) sensor possessing high selectivity and sensitivity was established for determination of lysozyme using magnetic graphene oxide–multi-walled carbon nanotube surface molecularly imprinted polymer (MGO–MWCNTs/SMIP). The MGO–MWCNTs/SMIP was characterized by X-ray diffraction (XRD), Fourier transform infrared (FT-IR) spectroscopy and scanning electron microscopy (SEM), and the maximum adsorption capacity of MGO–MWCNTs/SMIP to lysozyme was found to be 140 mg g⁻¹. The MGO–MWCNTs/SMIP was fixed into a glass tube and was connected to a chemiluminescent analyzer. Then a MGO–MWCNTs/SMIP-flow injection chemiluminescence (MGO–MWCNTs/SMIP-CL) sensor based on a luminol–NaOH–H₂O₂ CL system was established for the determination of lysozyme. The proposed sensor with high selectivity and sensitivity responded linearly to the concentration of lysozyme over the range 5.04 × 10⁻⁹ to 4.27 × 10⁻⁷ g mL⁻¹ and the detection limit was 1.90 × 10⁻⁹ g mL⁻¹ (3δ). The recoveries ranged from 98% to 111% when determining lysozyme in eggs and the result was satisfactory. The advantageous properties of sensor hold the potential to be applied in protein analysis, analogizing to biological analysis.

---

1. Introduction

As a basic enzyme that can hydrolyze mucopolysaccharides of pathogens,^1,2 lysozyme widely exists in most living bodies.³ Lysozyme plays an important role in human immune regulation and is closely related to human health⁴ with good antibacterial, anti-inflammatory and anti-virus activity.⁵ Today, lysozyme is mainly used in biochemical research and has multiple clinical applications, such as the treatment of herpes and warts.^6–9 So far, some methods, such as ion exchange¹⁰ and resonance light scattering,^11,12 have been used for the determination of lysozyme. However, these methods suffer from poor selectivity, low sensitivity, poor anti-interference ability and expensive equipment. Therefore, development of a novel method for lysozyme detection is very important. In recent years, surface molecules imprinting (SMIP) technique with good selectivity has received increasing attention. Molecular imprinting refers to the technology that prepares polymers with selective recognition ability for molecular targeting. Deng¹³ synthesized a copper ion selective membrane by surface-modified molecular imprinting and the adsorption selectivity of the ion imprinted membrane was increased. Zayats¹⁴ obtained maltose imprinting in hydrogels by surface molecule imprinting technique and the recognition of protein was tuned at the molecular level.

Flow injection (FI) is a rapid online analytical technique.¹⁵ Chemiluminescence (CL) technology has the advantages of high sensitivity and a wide linear range.¹⁶ FI-CL was established with a combination of these two methods. This method has the merits of high sensitivity and wide linear range,¹⁷ and it is widely used in biology,¹⁸ pharmacy,¹⁹ environmental science,²⁰ and many other areas. Because of poor selectivity, FI-CL could not be directly used for analyzing complex samples.

The combination of SMIP with high specific recognition and FI-CL with high-sensitivity in establishing a novel sensor has great significance. Interference from coexisting substances and complex sample matrices was eliminated. Thereby, the anti-jamming capability of surface molecular imprinting chemiluminescence (SMIP-CL) analysis was substantially strengthened.

In this work, using MGO–MWCNTs as the supporting material, the lysozyme MGO–MWCNTs/SMIP was obtained. Owing to the presence of MWCNTs, the specific surface area of GO was increased obviously and the number of sites of action was increased accordingly. In addition, the agglomeration of GO was reduced by the existence of MWCNTs. With the advantages of easy separation of magnetic Fe₃O₄ nanoparticles, high adsorption ability of GO–MWCNTs and excellent specificity recognition of SMIP, the overall performance of MGO–MWCNTs/SMIP was increased dramatically. Subsequently, the adsorption properties of the polymers were studied. Then, the MGO–MWCNTs/SMIP-CL sensor was constructed with high selectivity and high sensitivity and an FI-CL analytical sensor for determination of lysozyme was established.

2. Materials and methods

2.1 Reagents

All the chemicals were of analytical reagent grade unless otherwise stated. Double-distilled water was used throughout this work. Lysozyme, acrylamide and methacrylic acid were purchased from Sinopharm Chemical Reagent Company (China). MWCNTs were purchased from Beijing Daojin Technology Company (China). Graphite was purchased from Tianjin Hongyan Chemical Reagent (China). 3-(Methacryloyloxy) propyl trimethoxysilane (MPS), N,N,-methylene-bis-acrylamide, diethylaminoethyl methacrylate (DMAEMA), ethylene glycol dimethacrylate (EGDMA) and N,N,N,N,-tetramethylethylenediamine were purchased from Aladdin Reagent Company (China).

2.2 Apparatus

The IFFM-E flow injection CL analyzer was purchased from Xi'an Remex Electronic Instrument High-Tech Ltd. The schematic of the luminol–NaOH–H₂O₂ CL system used in this study is shown in Fig. 1. The FI-CL analyzer was equipped with an automatic injection system and a detector. All of the components were connected with the flow system using polytetrafluoroethylene tubing (0.8 mm i.d.). The CL signal was analyzed with a personal computer.

Fig. 1 The schematic of flow injection chemiluminescence system.

2.3 Preparation of MGO–MWCNTs/SMIP of lysozyme

GO was prepared according to the reported procedure from natural graphite powder by a modified Hummers' method.²¹ Firstly, 1.0 g of graphite powder was added into a 500 mL three-necked flask. Then, 200 mL of mixed acid solution (180 mL H₂SO₄ + 20 mL HNO₃ solution) was added into the flask, cooled by immersion in an ice bath and stirred for 0.5 h. Subsequently, 6.0 g of KMnO₄ was slowly added and the reaction was carried out for 2 h while the temperature was kept at 90 °C. And the condensed city water was continuously for 12 h. In the hood, the ice bath was then removed and H₂O₂ was slowly added dropwise until the reaction was completed. The solution was kept stirring until no gas generation. The final product was then centrifuged, washed twice with 30 mL of 0.2 mol L⁻¹ HCl solution and several times with 95% ethanol. Finally, the product was dried in vacuum and brown GO was obtained.

GO–MWCNTs composites were obtained by using a previously reported procedure²² with modification. 1.0 g of MWCNTs was dissolved in 70 mL of HNO₃ solution (65–68 wt%). The suspension was stirred and refluxed for 11 h and the temperature was kept below 75 °C. The product was washed with distilled water until the pH = 6.5. The product was dried under a vacuum for 10 h at 80 °C. 0.3 g of GO was stripped in 80 mL of ethanol–water mixture (v : v = 1 : 1) for less than 3 h by sonication. 0.4 g of acidified MWCNTs was added to the alcohol–water solution at room temperature. Then, the mixed solution was sonicated for 30 min. The solution was transferred to a 200 mL Teflon-lined stainless steel autoclave and maintained at 200 °C for 6 h. Thereafter, the product was cooled to room temperature. A black cylindrical product was achieved and washed seven times with ethanol. Finally, the product of GO–MWCNTs was obtained after drying at 60 °C under vacuum.

MGO–MWCNTs composites were synthesized by using a previously reported procedure.²³ 50 mg of GO–MWCNTs was dissolved in 30 mL of ultrapure water and ultrasonicated for 0.5 h to disperse. 50 mg of FeCl₃ and 35 mg of FeCl₂ were added into this solution and supplemented with 20 mL of ultrapure water. Under the protection of N₂, the solution was stirred and heated at 90 °C. NH₃·H₂O (28%) was added into the solution until the pH = 9.0 and the solution was heated with stirring for 0.5 h. The solution was separated with a magnet after heating and cooling to room temperature, washed twice with ethanol and dried in a vacuum oven.

In this experiment, 0.3 g of magnetic nanoparticles was added into a mixed solution (2 mL MPS and 20 mL anhydrous toluene). Under the protection of N₂, the solution was refluxed for 12 h. Finally, the product was collected by an external magnetic field, washed with ultrapure water and the silane modification product was obtained.

Subsequently, 16 mg of N,N-methylene bisacrylamide, 32.6 mg of acrylamide, 0.1 mL of methacrylate, 0.1 mL of DMAEMA, 0.1 mL of EGDMA, 32 mg of lysozyme and 25 mL of phosphate buffer solution (PBS, pH = 7.47, c: 0.01 mol L⁻¹) were added into a 250 mL iodine flask and sonicated. Then, 120 mg of MGO–MWCNTs was dissolved in 15 mL of ethanol and 5 mL of PBS solution by ultrasonication. Then, the two solutions were mixed quickly and shaken for 1 h at 25 °C, then 30 mg of (NH₄)₂S₂O₈, 0.4 mL of N,N,N,N,-tetramethylethylenediamine and 15 mg of FeSO₄ were added into this solution. Subsequently, the solution was shaken under nitrogen protection for 2 h at 25 °C. The product was washed twice with distilled water. In order to remove the unreacted monomers and the template molecule, the product was washed with NaCl (0.5 mol L⁻¹) solution under ultrasonication. In the next step, to remove excess NaCl, the product was washed twice with distilled water and dried at 60 °C.

The MGO–MWCNTs/SNIP was obtained in the same way without the addition of lysozyme. The preparation process for MGO–MWCNTs/SMIP is shown in Fig. 2.

Fig. 2 The preparation process for MGO–MWCNTs/SMIP.

2.4 Adsorption tests of MGO–MWCNTs/SMIP and MGO–MWCNTs/SNIP

MGO–MWCNTs/SMIP and MGO–MWCNTs/SNIP were added into the lysozyme solution for adsorption under the same conditions. The same volume of lysozyme solution was taken for testing at different times to determine the adsorption capacity. A series of lysozyme standard solutions with different concentrations were prepared, and the same amounts of SMIP and SNIP were added into the solutions for adsorption of the lysozyme.

2.5 Procedure for lysozyme determination

The schematic for the CL sensor is shown in Fig. 1 and specific experimental steps were as follows:

(1) Enrichment of lysozyme. Deputy pump (pump 1) stopped working. The injection valve was in the sampling position and the main pump (pump 2) transferred the lysozyme solution flowing through the column of MGO–MWCNTs/SMIP for 60 s. The lysozyme was absorbed by MGO–MWCNTs/SMIP.

(2) Removing of residual impurities. Pump 1 stopped working, the injection valve was at the sampling location and pump 2 transferred water to wash away other substances except for lysozyme.

(3) Determination of lysozyme. Pumps 2 and 1 were started and the injector valve was in the injection position. H₂O₂ and luminol solution flowed together in the column of MGO–MWCNTs/SMIP for 50 s and reacted with lysozyme, which generated a CL signal.

(4) Cleaning of the MGO–MWCNTs/SMIP column. Pump 1 stopped working and the injection valve was in the sampling position. The main pump transferred water through the column of MGO–MWCNTs/SMIP for 70 s. The residue was washed away from the column of MGO–MWCNTs/SMIP for the test again.

3. Results and discussion

3.1 Characterization

FT-IR spectra were recorded by Fourier transform infrared spectrometer (PerkinElmer, USA) with KBr pellets and were used to investigate the chemical groups on the surface of MGO–MWCNTs, MGO–MWCNTs/SMIP and MGO–MWCNTs/SNIP, and the results are shown in Fig. 3.

Fig. 3 The FT-IR spectra of MGO–MWCNTs (a), MGO–MWCNTs/SMIP (b) and MGO–MWCNTs/SNIP (c).

In the spectrum of MGO–MWCNTs, the peak at 3445 cm⁻¹ is the telescopic symmetrical characteristic peak of –OH. 1735 cm⁻¹ is the characteristic peak of carboxyl. The peak at 620 cm⁻¹ is the characteristic peak of Fe₃O₄, which gives evidence of the successful preparation of the MGO–MWCNTs. In the spectrum of MGO–MWCNTs/SMIP, peaks at 1385 cm⁻¹ (the characteristic peak of –CH₃), 1465 cm⁻¹ (the characteristic peak of –CH₂) and 1533 cm⁻¹ (the characteristic peak of –NH–) were able to justify the preparation of the SMIP. Compared to MGO–MWCNTs/SNIP, the characteristic peak at 3000 cm⁻¹ has an obvious displacement effect from hydrogen bonding. The results prove that hydrogen bonding exists in the synthesized SMIP.

X-ray diffraction (XRD) measurements were employed to investigate the phase and structures of GO, MWCNTs, MGO–MWCNTs and Fe₃O₄. As shown in Fig. 4a, the GO showed a characteristic peak at 2θ = 10.9°, showing that the GO was synthesized successfully.²⁴ Fig. 4b shows the typical XRD patterns of the MWCNTs and a sharp peak was located at 2θ = 26°. The peaks in Fig. 4d at 2θ values of 30.2°, 35.5°, 43.2°, 53.6°, 57.1° and 62.7° were consistent with the standard XRD data of Fe₃O₄. All characteristic peaks of GO, MWCNTs and Fe₃O₄ are included in Fig. 4c, which gives evidence of the successful preparation of the MGO–MWCNTs.

Fig. 4 XRD patterns of GO (a), MWCNTs (b), Fe₃O₄ (c) and MGO–MWCNTs (d).

Fig. 5 shows the SEM images of the obtained GO, GO–MWCNTs and MGO–MWCNTs. As shown in Fig. 5, the GO (a) presents a sheet-like structure, smooth surface, and wrinkled edge. In addition, GO and MWCNTs were uniformly intertwined, as shown in Fig. 5b. It proves that GO–MWCNTs were synthesized successfully and Fe₃O₄ had coated onto the GO–MWCNTs surface, as shown in Fig. 5c.²⁵ The MGO–MWCNTs/SMIP shown in Fig. 5d and MGO–MWCNTs/SNIP shown in Fig. 5e are nanoparticles and the roughness of the surface was different. The MGO–MWCNTs/SMIP contains cavities of lysozyme and the surface of the particles is rough, and MGO–MWCNTs/SNIP is smooth without imprinted cavities. The MGO–MWCNTs/SMIP was prepared with imprinted cavities.

Fig. 5 SEM images of the obtained GO (a), GO–MWCNTs (b), MGO–MWCNTs (c), MGO–MWCNTs/SMIP (d) and MGO–MWCNTs/SMIP (e).

The BET surface areas of MGO–MWCNTs/SMIP and MGO–MWCNTs/SMIP were determined from Barrett–Joyner–Halenda (BJH) analysis of the isotherms. The surface area of MGO–MWCNTs/SNIP was 24.214 m² g⁻¹ and the surface area of MGO–MWCNTs/SMIP was 57.295 m² g⁻¹. Obviously, the surface area of MGO–MWCNTs/SMIP was larger than that of MGO–MWCNTs/SNIP.

3.2 Adsorption study of MGO–MWCNTs/SMIP and MGO–MWCNTs/SNIP

The adsorption property of MGO–MWCNTs/SMIP to lysozyme is shown in Fig. 6a. As can be seen from the adsorption kinetics of MGO–MWCNTs/SMIP, lysozyme bound to the polymers quickly at the beginning of the adsorption. The adsorption amount of MGO–MWCNTs/SMIP increased and reached the maximum adsorption (140 mg g⁻¹) rapidly. At the beginning of the adsorption, the hole of the SMIP could quickly capture the lysozyme molecule. When most of the binding sites on the surface were occupied, the adsorption rate of SMIP to lysozyme decreased gradually owing to the large steric hindrance. Similarly, the amount of SNIP molecules adsorbed increased rapidly at the beginning stages of adsorption, but the adsorption capacity of SNIP was less than that of SMIP. The result can be explained by there being no specific recognition cavities for lysozyme in SNIP molecules.

Fig. 6 Adsorption dynamics curve of MGO–MWCNTs/SMIP and MGO–MWCNTs/SNIP (a); adsorption concentration curve of MGO–MWCNTs/SMIP and MGO–MWCNTs/SNIP (b).

Fig. 6b shows the adsorption capacity of MGO–MWCNTs/SMIP molecules and MGO–MWCNTs/SNIP molecules, and the adsorption capacity increased with the increasing concentration of lysozyme. In addition, when the lysozyme concentration reached a certain level, the adsorption capacity of the MGO–MWCNTs/SMIP and MGO–MWCNTs/SNIP would not increase. However, the adsorption capacity of MGO–MWCNTs/SNIP (95 mg g⁻¹) is much lower than the adsorption amount of MGO–MWCNTs/SMIP (140 mg g⁻¹). This is owing to there being no imprinted cavities in MGO–MWCNTs/SNIP, so only MGO and MWCNTs were involved in the adsorption of lysozyme. However, not only MGO and MWCNTs were involved in the adsorption of lysozyme in MGO–MWCNTs/SMIP; the MGO–MWCNTs/SMIP molecule also has a specific recognition cavity for lysozyme, which increases the adsorption capacity of SMIP further. Therefore, the differences between MGO–MWCNTs/SMIP and MGO–MWCNTs/SNIP in adsorption appeared. In addition, a large hydrophobic groove (the active site of lysozyme) is present on the surface of lysozyme and there are a large number of amino and carboxyl groups on the surface of lysozyme. Therefore, lysozyme can interact with many small molecules. Therefore, the nonspecific adsorption of lysozyme is serious and the adsorption capacity of MGO–MWCNTs/SMIP is only 50% higher than that of MGO–MWCNTs/SNIP.

3.3 Optimization of MGO–MWCNTs/SMIP-CL sensor

A diagram of FI-CL is shown in Fig. 1. The pump 2 speed, the pump 1 speed and the concentrations of the luminescent reagents were optimized.

The experimental results are shown in Fig. 7. The FI-CL signal was greatly affected by the concentration of H₂O₂ (Fig. 7a). With the H₂O₂ concentration in the range from 0.02 to 0.06 mol L⁻¹, the CL intensity reached the maximum at a 0.05 mol L⁻¹ concentration of H₂O₂. Thus, the optimum concentration of H₂O₂ was 0.05 mol L⁻¹.

Fig. 7 Optimization results. (a) Effect of H₂O₂ concentration on CL intensity. (b) Effect of NaOH concentration on CL intensity. (c) Effect of luminol concentration on CL intensity. (d) The regression equation of lysozyme. (e) Effect of main pump (pump 2) speed on CL intensity. (f) Effect of deputy pump (pump 1) speed on CL intensity.

The effect of the concentration of NaOH solution on the CL intensity was investigated in Fig. 7b. When the NaOH concentration was in the range of 0.01–0.24 mol L⁻¹, the CL intensity increased with the concentration of NaOH up to 0.03 mol L⁻¹. However, when the concentration of NaOH solution was over 0.03 mol L⁻¹, the CL intensity decreased. Thus, 0.03 mol L⁻¹ was the ideal choice for the concentration of NaOH solution.

The influence of the concentration of the luminol solution was examined over the range of 1.0 × 10⁻⁴ to 8.0 × 10⁻⁴ mol L⁻¹; the CL intensity reached maximum when the concentration of luminol was 6.0 × 10⁻⁴ mol L⁻¹, as shown in Fig. 7c.

The effect of the main pump (pump 2) speed on the luminous intensity is explored in Fig. 7e. As we can observe, the optimal speed of pump 2 was 45 rpm.

The effect of the deputy pump (pump 1) speed on CL intensity was studied, as shown in Fig. 7f. The optimal pump speed of pump 1 was 25 rpm.

3.4 Analytical performance of sensor

Under the optimum conditions, the linear response range of the sensor for lysozyme was obtained as shown in Fig. 7d. The linear equation is expressed as ΔI = −1.14 × 10² + 1.99 × 10² log c (c was the lysozyme concentration) and the correlation coefficient was 0.9950. The detection limit was 1.90 × 10⁻⁹ g mL⁻¹ and the sensor responded linearly to the lysozyme over the range 5.04 × 10⁻⁹ to 4.27 × 10⁻⁷ g mL⁻¹. Thence, it was proved that the method has a low detection limit.

The sensor was then placed in a vacuum oven. Two weeks later, the sensor was used to detect lysozyme and the sensor performance did not change significantly. The RSD was within the acceptable range. Therefore, the sensor exhibited good stability.

3.5 Interferences study

Under the explored conditions (luminol: 6.0 × 10⁻⁴ mol L⁻¹, NaOH: 0.03 mol L⁻¹, H₂O₂: 0.05 mol L⁻¹, pump 2 speed: 45 rpm, pump 1 speed: 25 rpm), coexisting substances were added to a lysozyme solution (3.0 × 10⁻⁸ mol L⁻¹) to investigate the effect on CL intensity. Various substances were added into lysozyme solution to examine the effects on the determination of lysozyme. The tolerance for interfering substances in samples with MGO–MWCNTs/SMIP and MGO–MWCNTs/SNIP columns was compared when the relative error was less than ±5% and the tolerance times are shown in Fig. 8.

Fig. 8 Interference study. (1) Fe³⁺; (2) Na⁺; (3) citric acid; (4) cytochrome C; (5) lactate; (6) bovine serum albumin; (7) bovine hemoglobin.

These results show that when MGO–MWCNTs/SMIP was used as the sensor, the detection of lysozyme was not affected by 400 times the concentration of Fe³⁺, 520 times the concentration of Na⁺, 500 times the concentration of citric acid, 100 times the concentration of cytochrome C, 480 times the concentration of lactate, 130 times the concentration of bovine serum albumin and 50 times the concentration of bovine hemoglobin. The anti-jamming capability of the sensor was increased obviously. Thus, the sensor could be used for lysozyme analysis and the selectivity has been increased dramatically.

3.6 Application of MGO–MWCNTs/SMIP-CL sensor

The sensor was used to detect lysozyme in eggs. Egg samples need to be processed before analysis: the egg was cracked and the egg white was diluted with PBS (pH = 7.47, c: 0.01 mol L⁻¹). The solution was bathed at 77 °C for 10 min and was centrifuged in order to obtain the supernatant. Then, the supernatant was diluted tenfold. According to the test method, the spiked recovery experiment was carried out.

The results are shown in Table 1. The recoveries varied from 98% to 111%. As a result, the CL sensor used for the determination of lysozyme was practical.

Table 1 Determination results of samples (n = 6)

Samples	Content (ng mL^−1)	Added (ng mL^−1)	Found (ng mL^−1)	Recovery (%)	RSD (%)
1^#	200	100	311	111	3.4
2^#	100	100	198	98	3.7
3^#	50	100	156	106	4.1

---

3.7 Possible mechanism of the reaction

The structure of lysozyme consists of α-helix, β-fold, β-corner and random coil. Lysozyme is a globular protein containing 129 sequences of amino acid residues with the active center in the cleft between the two domains on the molecule surface.²⁶ Six tryptophan residues are located at the binding site of lysozyme and the 62nd and 108th tryptophan residues are the main fluorescent groups.

In view of the widespread use of H₂O₂ in protein modification, the mechanism of H₂O₂ oxidizing certain amino acids in lysozyme is not clear up to now. However, Song²⁷ proved that lysozyme can enhance the chemiluminescence signal and the possible mechanism of the reaction was proposed. Therefore, it may be concluded that the enhancement mechanism is presented below:

Tryptophan (in lysozyme) + H₂O₂ + OH⁻ → intermediate radical (in lysozyme) (1)

Intermediate radical (in lysozyme) + luminol → oxidation products + hν (2)

4. Conclusions

In this work, MGO–MWCNTs/SMIP, which exhibited high selectivity for lysozyme, was synthesized. The MGO–MWCNTs/SMIP was characterized by SEM, XRD and FT-IR. The adsorption properties of the polymer were studied. Then, a new CL method for the determination of lysozyme based on SMIP was achieved. A luminol–NaOH–H₂O₂ CL system was selected and the optimum conditions for CL were explored. The proposed method responded linearly to the concentration of lysozyme over the range was 5.04 × 10⁻⁹ to 4.27 × 10⁻⁷ g mL⁻¹. The detection limit was 1.90 × 10⁻⁹ g mL⁻¹ (3δ), which reflected that the sensor was satisfactory. The advantageous properties of the sensor hold the potential to be applied in protein analysis, analogizing to biological analysis.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (NSFC, No. 21345005 and 21205048), the Shandong Provincial Natural Science Foundation of China (No. ZR2012BM020) and the Scientific and Technological Development Plan Item of Jinan City in China (No. 201202088).

References

1. W. Mai and C. Hu, cDNA cloning, expression and antibacterial activity of lysozyme C in the blue shrimp (Litopenaeus stylirostris), Prog. Nat. Sci., 2009, 19, 837–844.
2. A. Cheng, B. Ge and H. Yu, Aptamer-based biosensors for label-free voltammetric detection of lysozyme, Anal. Chem., 2007, 79, 5158–5164.
3. N. Sun, S. Chen, X. Xie, Y. Wang and G. Li, et al., Expression, characterization and antimicrobial ability of a variant T4 lysozyme in Pichia pastoris, Agric. Sci. Technol., 2014, 15, 321–325.
4. Q. Ren, X. Zhao and J. Wang, Molecular characterization and expression analysis of a chicken-type lysozyme gene from housefly (Musca domestica), J. Genet. Genomics, 2009, 36, 7–16.
5. N. Wang, Y. Wang, G. Li and N. Sun, Expression, characterization and antimicrobial ability of T4 lysozyme from methylotrophic yeast Hansenula polymorpha A16, Sci. China: Life Sci., 2011, 54, 520–526.
6. S. Morita, M. Kuriyama, M. Nakatsu, M. Suzuki and K. Kitano, Secretion of active human lysozyme by acremonium chrysogenum using a fusarium alkaline protease promoter system, J. Biotechnol., 1995, 42, 1–8.
7. E. Osserman, M. Klockars, J. Halper and E. Fischel, Effects of lysozyme on normal and transformed mammalian cells, Nature, 1973, 243, 331–335.
8. J. Warren, J. Rinehart, B. Zwilling and J. Neidhart, Lysozyme enhancement of tumor cell immunoprotection in a murine fibrosarcoma, Cancer Res., 1981, 41, 1642–1645.
9. R. Wooley, W. Schall and R. Eagon, Efficacy of EDTA-Tris-lysozyme lavage in the treatment of experimentally in induced pseudomonas aeruginosa cystitis in the dog, Am. J. Vet. Res., 1974, 35, 27–29.
10. T. Durance and D. Timothy, Process for isolation of lysozyme and avidin from egg white, US 4, 996, 851, October 30, 1990.
11. R. Pasternack and P. Collings, Resonance light-scattering-a new technique for studying chromophore aggregation, Science, 1995, 269, 935–939.
12. R. Pasternack, C. Bustamante, P. Collings, A. Giannetto and E. Gibbs, Porphyrin assemblies on DNA as studied by a resonance light-scattering technique, J. Am. Chem. Soc., 1993, 115, 5393–5399.
13. H. Deng, L. Gao, S. Zhang and J. Yuan, Preparation of a copper ion selective membrane by surface-modified molecular imprinting, Ind. Eng. Chem. Res., 2012, 51, 14018–14025.
14. M. Zayats, M. Kanwar, M. Ostermeier and P. Searson, Molecular imprinting of maltose binding protein: tuning protein recognition at the molecular level, Macromolecules, 2011, 44, 3966.
15. H. Behzad and B. Somayyeh, Flow injection chemiluminescence determination of isoniazid using luminol and silver nanoparticles, Microchem. J., 2010, 95, 192–197.
16. K. Tsukago, M. Sumiyama and J. Naka, Chemiluminescence property of the luminol–hydrogen peroxide–copper system in the presence of surface-carboxylated microspheres, Anal. Sci., 1998, 14, 409–412.
17. B. Rezaei and A. Mokhtari, A simple and rapid flow injection chemiluminescence determination of cysteine with Ru(phen)²⁺–Ce(rV) system, Spectrochim. Acta, Part A, 2007, 66, 359–363.
18. J. Liu, L. Zhang, Y. Wang, Y. Zheng and S. Sun, An improved portable biosensing system based on enzymatic chemiluminescence and magnetic immunoassay for biological compound detection, Measurement, 2014, 47, 200–206.
19. H. Yao, E. Xu, W. Zeng, X. Zeng and M. Zhang, et al., Determination of doxorubicin in pharmaceutical preparation and rat plasma with luminol–K₃Fe(CN)₆ chemiluminescence system, J. Food Drug Anal., 2013, 21, 279–285.
20. X. Tan, Z. Wang, D. Chen, K. Luo and Z. Song, et al., Study on the interaction of catalase with pesticides by flow injection chemiluminescence and molecular docking, Chemosphere, 2014, 108, 26–32.
21. W. Hummers and R. Offeman, Preparation of graphitic oxide, J. Am. Chem. Soc., 1958, 80, 1339.
22. X. Qin, H. Wang, X. Wang, Z. Miao and L. Chen, et al., Amperometric biosensors based on gold nanoparticles-decorated multiwalled carbon nanotubes–poly(diallyldimethylammonium chloride) biocomposite for the determination of choline, Sens. Actuators, B, 2010, 147, 593–598.
23. H. Qiu, C. Luo, M. Sun, F. Lu and L. Fan, et al., A chemiluminescence array sensor based on graphene–magnetite-molecularly imprinted polymers for determination of benzenediol isomers, Anal. Chim. Acta, 2012, 744, 75–81.
24. D. Han, L. Yan, W. Chen and W. Li, Preparation of chitosan/graphene oxide composite film with enhanced mechanical strength in the wet state, Carbohydr. Polym., 2011, 83, 653–658.
25. H. Qiu, C. Luo, M. Sun, F. Lu and L. Fan, et al., Determination of L-tryptophan based on grapheneoxide–magnetite-molecularly imprinted polymers and chemiluminescence, Talanta, 2012, 98, 226–230.
26. Z. Wang and Z. Song, A valuable way for understanding the relationships between lysozyme and cephalosporin analogues by flow injection chemiluminescence, Analyst, 2010, 135, 2546–2553.
27. Z. Song and S. Hou, A new analytical procedure for assay of lysozyme in human tear and saliva with immobilized reagents in flow injection chemiluminescence system, Anal. Sci., 2003, 19, 347–352.

---