Journal of Chromatography A, 1340 (2014) 139–145
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Screening of neuraminidase inhibitors from traditional Chinese medicines by integrating capillary electrophoresis with immobilized enzyme microreactor Haiyan Zhao, Zilin Chen ∗ Key Laboratory of Combinatorial Biosynthesis and Drug Discovery (Wuhan University), Ministry of Education, and Wuhan University School of Pharmaceutical Sciences, Wuhan 430071, China
a r t i c l e
i n f o
Article history: Received 30 January 2014 Received in revised form 5 March 2014 Accepted 10 March 2014 Available online 18 March 2014 Keywords: Neuraminidase Inhibitor screening Immobilized enzyme microreactor Short-end injection Traditional Chinese medicine Capillary electrophoresis
a b s t r a c t A simple and effective neuraminidase-immobilized capillary microreactor was fabricated by glutaraldehyde cross-linking technology for screening the neuraminidase inhibitors from traditional Chinese medicines. The substrate and product were separated by CE in short-end injection mode within 2 min. Dual-wavelength ultraviolet detection was employed to eliminate the interference from the screened compounds. The parameters relating to the separation efficiency and the activity of immobilized neuraminidase were systematically evaluated. The activity of the immobilized neuraminidase remained 90% after 30 days storage at 4 ◦ C. The immobilized NA microreactor could be continuously used for more than 200 runs. The Michaelis–Menten constant of neuraminidase was determined by the microreactor as 136.6 ± 10.8 M. In addition, six in eighteen natural products were found as potent inhibitors and the inhibition potentials were ranked in the following order: bavachinin > bavachin > baicalein > baicalin > chrysin and vitexin. The half-maximal inhibitory concentrations were 59.52 ± 4.12, 65.28 ± 1.07, 44.79 ± 1.21 and 31.62 ± 2.04 for baicalein, baicalin, bavachin and bavachinin, respectively. The results demonstrated that the neuraminidase-immobilized capillary microreactor was a very effective tool for screening neuraminidase inhibitors from traditional Chinese medicines. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Influenza virus infection is a major public health problem worldwide, causing serious respiratory illnesses and high death toll each year. Neuraminidase (EC 3.2.1.18, NA) cleaves the specific linkage of the sialic acid receptor to facilitate the release of influenza virus to infect new cells [1]. Additionally, NA may promote the early process of influenza virus infection of lung epithelial cells [2]. Due to the key role in the replication, spread, infection and pathogenesis of influenza virus, the application of neuraminidase inhibitors (NAIs) has been considered as a dominant approach for the treatment of influenza infection. In particular, NAIs are effective for influenza type A and B viruses in human, avian and animal. Among the available neuraminidase inhibitors, oseltamivir and zanamivir are the most widely used drugs. However, their applications are limited due to the rapid emergence of increasing numbers of resistant
∗ Corresponding author at: School of Pharmaceutical Sciences, Wuhan University, Wuhan 430071, China. Tel.: +86 27 68759893; fax: +86 27 68759850. E-mail address: [email protected] (Z. Chen). http://dx.doi.org/10.1016/j.chroma.2014.03.028 0021-9673/© 2014 Elsevier B.V. All rights reserved.
strains during their clinical treatment [3,4], therefore it is necessary to develop novel and effective antiviral agents. A variety of methods have been developed for the discovery of novel NAIs, including NA enzyme-based inhibitory assays, cell culture-based screening, X-ray crystal structure-based computational docking studies and virtual screening [5]. The enzyme-based inhibition assay is the most widely used method by either fluorescence (FL) [6,7] or chemiluminescence (CL) detection [8,9]. However, CL has some disadvantages. For instance, color quenching and cross talk may lead to significant errors. In FL detection, the fluorogenic substrate may nonlinearly interfere with the fluorescence measurement of its product 4-methylumbelliferone (4-MU). The autofluorescence substances such as flavonoids have been reported to exert a considerable quenching effect causing falsepositive results [10]. To solve these problems, a correction factor was introduced according to the fluorescence measurements of 4-MU and 4-MU mixed with substrate [11]. Recently, simple and sensitive affinity screening methods based on MS have been applied to screen potential NAIs in the absence of substrate. Liu et al. developed an HPLC-electrospray ionization tandem mass spectrometric method combined with ultrafiltration for studying NA inhibitory
140
H. Zhao, Z. Chen / J. Chromatogr. A 1340 (2014) 139–145
activity of n-butanol extract of Compound Indigowoad Root Granule [12]. Matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS) has been used to identify the active site domains within influenza NA and to screen inhibitors by control measurements of peptide residues of NA alone and NA-inhibitor complexes after tryptic digestion [13–15]. Among the reported methods, the enzymic reaction was carried out by off-line incubation for 0.5–1 h, which was time-consuming and sample-consuming. As an alternative, online incubation with immobilized enzyme attracts much interest. NA and the substrate were immobilized onto a combinable sensor array for inhibitors assay [16]. The sensor simplified the conventional multiple steps into a single-step operation. However, the method suffers from the enzymic deactivation in three days and the interference from the fluorogenic substrate. The immobilized enzyme microreactor (IMER) has many remarkable advantages such as automatic operation, compatibility with detectors, low consumption of samples, high enzyme activity and reusability. The development and use of IMERs have recently been reviewed [17]. This work aims to develop a new NA-immobilized capillary microreactor integrated with capillary electrophoresis to screen potential NA inhibitors from traditional Chinese medicines. Capillary electrophoresis was employed to separate the substrate and product with diode array detection (DAD). The interference from the screened compounds was eliminated by dual-wavelength detection. To enhance the screening efficiency, a short-end injection procedure [18,19] (sample injected from the capillary outlet end) was introduced by applying a negative pressure at the inlet. The parameters relating to the separation efficiency and NA activity were optimized and the kinetics constant was tested under the optimal experimental conditions. Finally, this method was further verified by eighteen compounds from traditional Chinese medicines. 2. Materials and methods 2.1. Chemicals The substrate 2 -(4-methylumbelliferyl)-␣-d-N-acetylneuraminic acid sodium salt hydrate (4-MuNeu5Ac), the product 4-methylumbelliferone (4-MU) and neuraminidase (10 UN in package, from Clostridium perfringens, Type V, lyophilized powder) were purchased from Sigma–Aldrich (Steinheim, Germany). 3-Aminopropyltriethoxysilane (APTES) was obtained from Aladdin Reagent Co., Ltd. (Shanghai, China). Glutaraldehyde solution in water (25%, m/v) (GA) and some analytical reagents such as sodium dihydrogen phosphate, disodium hydrogen phosphate, sodium acetate anhydrous and acetic acid were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All the screened compounds were obtained from Shanghai Shunbo Bio-engineering Technology (Shanghai, China). Deionized water was purified using a Milli-Q System (Millipore, Bedford, MA, USA). Other reagents used in the experiment were of analytical grade and from commercial sources. 2.2. Instrumentation The capillary electrophoresis system was an Agilent 7100 CE system (Waldbronn, Germany) equipped with a photodiode array detector. CE separations were performed in fused-silica capillary (35.5 cm total length, 8.5 cm from the outlet end to the detection window, 50 m i.d., 365 m o.d.) obtained from Ruifeng Chromatographic Devices (Yongnian, Hebei, China). New capillaries were preconditioned by 1.0 M NaOH for 2 h and deionized water for 0.5 h. After injection into the capillary outlet end by applying a negative
pressure of −5 kPa for 6 s at the inlet, the analytes were separated at +20 kV with a BGE of 10 mM sodium phosphate buffer (pH 5.5). For inhibition assay, baicalein, baicalin and vitexin were detected at 320 nm with a reference wavelength of 260 nm. The other compounds were directly detected at 320 nm with a bandwidth 4 nm. The capillary was thermostatted at 30 ◦ C. Between runs, the capillary was flushed with water and BGE at 95 kPa for 2 min and 5 min, respectively.
2.3. Solution preparation The BGE was prepared by adding 10 mM disodium hydrogen phosphate to 10 mM sodium dihydrogen phosphate under continuous stirring till pH value reached 5.5. Stock solutions of substrate 4-MuNeu5Ac and NA were prepared in deionized water at the concentration of 2 mM and 10 U/mL, respectively. Then the stock solutions were divided into small quantities and stored at −20 ◦ C. Product 4-MU and the screened compounds were dissolved in methanol and stored at 4 ◦ C. The substrate, NA and the screened compounds were separately diluted with pH 5.5 phosphate buffer prior to use each day. All the prepared solutions were filtered by 0.45 m nylon membrane.
2.4. Preparation of immobilized NA capillary microreactor NA was covalently immobilized onto the capillary wall as the previously reported method [20–22] with some modifications. The preparation protocol is shown in Fig. 1. Briefly, the inner surface of the capillary was first coated with APTES by pumping the solution of 30% (v/v) APTES in methanol through the capillary for 5 min. Two ends of the capillary were connected by a piece of Teflon tube and polymerized at 95 ◦ C for 1 h. The capillary was washed with methanol and dried by nitrogen stream. Next, a glutaraldehyde solution 2.5% (v/v) in pH 7.0 sodium phosphate buffer was introduced into the capillary outlet end by capillary wetting to make a 5 cm length coating in the capillary outlet end. Two ends of the capillary were connected by a piece of Teflon tube and reacted in 40 ◦ C water bath for 2 h. Thus the amine groups of APTES bonded to aldehyde group on one side of GA. The capillary was then backflushed with pH 7.0 sodium phosphate buffer for 30 min to remove redundant GA. After dried by nitrogen, the capillary was treated with 0.5 U/mL NA in pH 7.0 sodium phosphate buffer and then incubated at 25 ◦ C for 30 min. Unreacted enzyme was back-flushed out with phosphate buffer, then the capillary was stored at 4 ◦ C for 24 h. Thus, NA was covalently immobilized onto the capillary outlet end by the coupling of amino groups of NA and aldehyde groups on the other side of GA. Before use, the extra 4 cm length of the capillary outlet was cut off to leave 1 cm in length of the immobilized enzyme microreactor. While not in use, the capillary was maintained in pH 7.0 sodium phosphate buffer and stored at 4 ◦ C.
2.5. Kinetics study of NA 1-cm length of substrate plug was injected into the capillary from the outlet end and incubated for 1 min. The substrate and product were separated at +20 kV with BGE of 10 mM sodium phosphate buffer (pH 5.5). Following the on-line enzymic reaction, the rate of hydrolysis of the substrate was calculated according to the increase in the peak area of 4-MU with increasing substrate concentrations at 15, 31, 62.5, 125, 250, 500 and 1000 M. Each concentration was analyzed three times. The data were fitted to double reciprocal plot by Origin Pro 8.0 software (OriginLab, Northampton, MA, USA).
H. Zhao, Z. Chen / J. Chromatogr. A 1340 (2014) 139–145
141
Fig. 1. Procedure for immobilizing NA on capillary inner wall (a) and schematic of on-line enzyme assay (b).
2.6. Inhibitors screening A mixture of substrate and the test compound was diluted with BGE to the final concentration of 125 M and 60 M, respectively. The mixture was injected into the enzyme microreactor at −5 kPa for 6 s. After incubation for 4 min, the product was separated from the unreacted substrate under the optimal conditions, and each inhibitor assay was performed in triplicate (n = 3). Before each analysis, the capillary was flushed with water for 2 min and conditioned with BGE for 5 min. The inhibitors were identified by the reduction of the peak area of the product. The percentages of inhibition were calculated by the equation: inhibition% =
1−
AI AS
× 100%
(1)
at pH < 6.0 (isoelectric point of APTES at pH 6.0) [23], the surface of the modified capillary inner wall was positively charged, which resulted in a reversed EOF from the cathode to the anode. Hence, the separation voltage was set at +20 kV at pH of 3.0–5.5 and −20 kV at pH 6.0 for the injection from the outlet end. As shown in Fig. 2, the peak area of 4-MU and the resolution increased as the increase of pH. The analytes could be well separated at the all tested pH. The migration time of analytes was much dependent on the pH of BGE. The lower the BGE pH, the shorter was the migration time. This was due to the fact that the protonations of amino groups in APTES at lower pH would give rise to higher effective mobility and faster electrophoretic migration of the analytes. Because of its pKa value 7.8 [24], 4-MU is electric neutral and migrated with the same velocity as the EOF at the investigated pH range. Since the substrate was negatively charged at pH < 6.0, it electrophoretically migrated
where AI and AS represent the peak area of 4-MU obtained in the presence and absence of inhibitor, respectively. Half-inhibitory concentration (IC50 ) was determined at constant substrate concentration of 125 M with eight concentrations of inhibitors ranged from 2 to 320 M. Each concentration of inhibitor was assayed in triplicate. The dose-response curve for enzyme inhibition was plotted by fitting the percentage of enzyme activity as a function of log [inhibitor]. The enzyme activity (%) was defined as the ratio of the peak area of 4-MU in the presence and absence of the inhibitor. IC50 was defined as the inhibitor concentration necessary to reduce the enzyme activity by 50%. The data processing and curve-fitting were performed by PRISM® 5.0 (GraphPad software, San Diego, CA). 3. Results and discussion 3.1. Optimization of CE separation conditions The substrate and product were separated by capillary electrophoresis in fused silica capillary with modification of APTES. The BGE pH was optimized on the basis of the effect of pH (from 3.0 to 6.0) on the migration time and peak area of 4-MU (Fig. 2). Since the primary amino groups of APTES were protonated to be cationic
Fig. 2. Effect of BGE pH on the separation efficiency of substrate and product. CE conditions: fused-silica capillary with modification of APTES, 50 m i.d. × 35.5 cm (8.5 cm from the outlet end to the detector); BGE, 10 mM sodium phosphate buffer; separation voltage, +20 kV for pH 3.0–5.5 and −20 kV for pH 6.0; sample injection, −5 kPa, 6 s; UV detection at 320 nm; cartridge temperature, 30 ◦ C; concentration of substrate and product, 125 and 200 M, respectively.
142
H. Zhao, Z. Chen / J. Chromatogr. A 1340 (2014) 139–145
Fig. 3. Effect of incubation time (a and b), pH of sample solution (c), and the percentage of methanol (d) on the peak area of 4-MU by on-line enzymic reaction. CE conditions: fused-silica capillary with immobilized enzyme microreactor, 50 m i.d. × 35.5 cm (8.5 cm from the outlet end to the detector); sample injection, −5 kPa, 6 s; incubation time, 4 min; BGE, 10 mM pH 5.5 sodium phosphate buffer; separation voltage, +20 kV; UV detection at 320 nm; cartridge temperature, 30 ◦ C; concentration of substrate, 125 M.
in the same direction as EOF and faster than the product. At pH 6.0, the substrate electrophoretically migrated in the reverse direction of EOF and faster than EOF, therefore it would fall back into the outlet vial. To achieve shorter migration time and higher peak area of 4-MU, pH 5.5 was chosen as the optimum pH of BGE. To shorten the analytical time, the routine injection from the capillary inlet end (long-end injection mode) was replaced by the injection from the outlet end of the capillary (short-end injection mode). It has been demonstrated that outlet injection can greatly reduce the analysis time and enhance the sensitivity of detection [25]. In this study, the capillary effective length was shortened to 8.5 cm and the separation current was maintained. As shown in Fig. S1, the migration time of the product in long-end injection mode was four times longer than that in short-end injection mode.
3.2. Optimization of online immobilized NA reaction conditions It is necessary for the substrate to online incubate with the immobilized NA for several minutes to generate adequate product. The incubation time varied from 50 s to 600 s was used to examine the effect of incubation time on the peak area of 4-MU. The peak area of 4-MU increased linearly over the first 4 min with a determination coefficient of 0.9949 (Fig. 3a–b) but increased slightly after 4 min, which indicated that the incubation for 4 min was sufficient for the enzymic reaction. Thus, the incubation for 4 min was chosen to carry out the further experiments. To investigate the pH effect of sample solution on the reaction efficiency, the substrate was diluted by buffers with different pH values. The peak areas of the product showed no significant difference in pH span 3.0–7.0 whereas dramatically decreased at pH 8.0 (Fig. 3c), which indicated that pH 3.0–7.0 was feasible for additional enzymic reaction. In many investigations, pH 4.2–5.0 was commonly used for the enzymic reaction [26,27]. Additionally, the enzyme immobilization failed at pH value lower than 5.0 in this
study. Taking overall consideration of the IMER stability, the BGE for separation and the immobilized enzyme activity, pH 5.5 was selected as the optimal reaction condition. The effect of methanol on the reaction efficiency was evaluated by the substrate dilution with buffer containing different percentages of methanol. The peak area of 4-MU decreased slightly with the increase of methanol from 0 to 30% but significantly with the increase of methanol higher than 30% (v/v) (Fig. 3d). To avoid false results in inhibition assay, the final methanol concentration in the sample should be lower than 10% (v/v). 3.3. Method validation The reproducibility of the immobilization procedure was assessed by measuring the peak areas of 4-MU from enzymic reactions of six similar microreactors at fixed substrate concentration of 125 M. Each microreactor was assayed in five times. The RSD for the migration time and peak area of the product were 3.71% and 5.33%, respectively. The storage stability was examined by measuring the peak area of the product after storage at 4 ◦ C in BGE for 30 days, and the activity of immobilized NA retained 90% with RSD of 3.81%. Furthermore, the immobilized NA microreactor could be continuously used for at least 200 runs. The results showed that the microreactor had good stability. The intra-day and inter-day precision was estimated by five runs of on-line enzymic reaction at fixed substrate concentration of 125 M on the same day and on five consecutive days, respectively. The results showed that the RSD of the peak area of the product was 3.33% for intra-day precision and 6.15% for inter-day precision. 3.4. Kinetics study of immobilized NA For enzyme-catalyzed reactions, the relation between initial reaction velocity (V0 ) and substrate concentration [S] can be
H. Zhao, Z. Chen / J. Chromatogr. A 1340 (2014) 139–145
143
Fig. 4. Michaelis–Menten plot (a) and double reciprocal plot (b) at varied concentrations of substrate ranging from 15 to 1000 M by online enzymic reaction. Incubation time, 1 min; other CE conditions as in Fig. 3.
described by the Michaelis–Menten equation, and the kinetics parameters Km and Vmax can be calculated by linear regression from Lineweaver–Burk equation: Km 1 1 1 = + Vmax [S] Vmax V0
(2)
where Vmax is the maximum reaction velocity and Km is the Michaelis–Menten constant, the substrate concentration at half of the maximum velocity. To determine the kinetic constant (Km ), the conversion rate of the substrate should be less than 10%, hence incubation time was set at 1 min. The peak area of the product was used to indicate the initial reaction velocity (V0 ). After the online enzymic reaction, the Km was estimated by linear regression from double reciprocal plots with substrate concentrations ranging from 15 to 1000 M (Fig. 4). Each concentration was analyzed in triplicate (n = 3). The linear regression equation of the double reciprocal curve was y = 15.68x + 0.1148 with a determination coefficient of 0.9925. From the slope and the intercept on the ordinate, the Km value for the immobilized NA was calculated to be 136.6 ± 10.8 M which was in the range of 102–150 M reported in literatures [6,28,29]. 3.5. Inhibition assay by online NA-immobilized capillary microreactor Initially, the substrate, 4-MU and the screened compounds were separated by capillary electrophoresis under the optimal conditions. Baicalein and vitexin had the same migration time as 4-MU and baicalin had the same migration time as the substrate (Fig. S2a), which suggested that the three compounds may interfere with the detection of substrate and 4-MU. Dual-wavelength detection was adopted to eliminate this interference by setting the equal absorbance wavelength of baicalein, baicalin and vitexin as the reference wavelength in DAD. Since the absorbances of the three compounds at 320 nm were equal to those at 260 nm whereas 4MU had scarce absorption at 260 nm (Fig. S3), 260 nm was chosen as the reference wavelength (see the supporting information). As shown in Fig. S2b, the peaks of the three compounds disappeared, while the peak height of the product remained almost unchanged. Under the optimal experimental conditions, the NAimmobilized microreactor was applied to investigate the inhibition potential of eighteen natural products. The stock solutions of these compounds were individually mixed with the substrate stock solution and diluted by the BGE to achieve the final concentration of the substrate at 125 M and the screened compounds at 60 M. These compounds were kept at constant concentration for ranking
Table 1 Inhibitors screening by immobilized neuraminidase microreactor (n = 3). Compounds
I%
Baicalein Baicalin Vitexin Chrysin Bavachin Bavachinin Resveratrol Psoralen Isopsoralen
62.4 ± 56.7 ± 42.3 ± 43.5 ± 81.1 ± 85.1 ± 0% 0% 0%
5.1% 3.5% 3.5% 2.3% 7.8% 6.4%
Compounds
I%
Oleanolic acid Ursolic acid Demethoxycurcumin Lycorine Galanthamine Matrine Oxymatrine Aconitine Vinblastine
0% 0% 0% 0% 0% 0% 0% 0% 0%
their inhibition potential. The overlaid electropherograms of online enzymic reaction in the presence of inhibitors are shown in Fig. 5 and the percentages of inhibition are summarized in Table 1. The results indicated that six natural products showed potential inhibition for NA. The inhibition potentials were ranked in the following order: bavachinin > bavachin > baicalein > baicalin > chrysin and vitexin. The ranking order of baicalein, baicalin, chrysin and vitexin was consistent with that in the literature [30]. Structure–activity study demonstrated that the inhibition activity would reduce when hydroxyl groups on C-5 or C-7 position were replaced by glycosyl groups, whereas remained unchanged when replaced by methoxyl groups. Due to the lack of C4 -OH, chrysin and vitexin displayed weaker activity [30]. The inhibition effects of bavachinin and bavachin were firstly investigated in our work. The dose-response curves for four potent inhibitors showed that these compounds had dose-dependent inhibitory effects against immobilized NA (Fig. 6). The enzyme activity decreased with the increasing of the inhibitor concentrations, which indicated that the enzyme activity was inhibited in the capillary after online enzymic reaction. The IC50 values were determined as 59.52 ± 4.12, 65.28 ± 1.07, 44.79 ± 1.21 and 31.62 ± 2.04 for baicalein, baicalin, bavachin and bavachinin, respectively. Due to their poor solubility, vitexin and chrysin were dissolved in high percentage of methanol which would lead to false-positive results. Therefore, the IC50 values of vitexin and chrysin were not determined. The IC50 values of baicalin and baicalin were higher than those reported in the literature [12]. The difference can be explained by the fact that the concentrations of the inhibitors increased after ultrafiltration [12]. The magnitude of IC50 depends upon the concentration of the substrate, origin of the enzyme and the assay conditions. Hence, the absolute value of IC50 is less important than the order of magnitude and the relative value between two inhibitors [31,32].
144
H. Zhao, Z. Chen / J. Chromatogr. A 1340 (2014) 139–145
Fig. 5. The overlaid electropherograms for inhibitors screening by online immobilized enzymic assay detected at 320 nm with a reference wavelength of 260 nm (a) and at 320 nm without reference wavelength (b). CE conditions as in Fig. 3.
Fig. 6. Dose-response curves fit of neuraminidase activity (%) as a function of log [inhibitor]. Each point represents average activity and error bars indicate the SD; n = 3.
4. Conclusions The first online NA inhibitor screening method was successfully established by CE based on immobilized NA capillary microreactor. The method was effective and low cost because of the reusable enzyme and the nanoliter level consumption of the samples. The integration of enzymic reaction and separation into one capillary rendered the inhibitors screening fully automated. The short-end injection mode greatly enhanced the screening efficiency. Additionally, the dual-wavelength detection eliminated the interference of the screened compounds and minimized the
false-positive results. The inhibitors screening method based on immobilized NA microreactor developed in this study provided a framework for discovering new NAIs.
Acknowledgements This work was supported by the National Natural Science Foundation of China (Grant nos. 21375101 and 30973672), Wuhan Science and Technology Bureau (No. 20140601010057), Doctoral Fund of Ministry of Education of China (No. 20110141110024),
H. Zhao, Z. Chen / J. Chromatogr. A 1340 (2014) 139–145
Innovation Seed Fund and Translational Medical Research Fund of Wuhan University School of Medicine. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.chroma.2014. 03.028. References [1] G. Neumann, T. Noda, Y. Kawaoka, Nature 459 (2009) 931. [2] M.N. Matrosovich, T.Y. Matrosovich, T. Gray, N.A. Roberts, H.D. Klenk, J. Virol. 78 (2004) 12665. [3] A.J. Hay, F.G. Hayden, Lancet 381 (2013) 2230. [4] Y. Matsuzaki, K. Mizuta, Y. Aoki, A. Suto, C. Abiko, K. Sanjoh, K. Sugawara, E. Takashita, T. Itagaki, Y. Katsushima, Virol. J. 7 (2010) 53. [5] U. Grienke, M. Schmidtke, S. von Grafenstein, J. Kirchmair, K.R. Liedl, J.M. Rollinger, Nat. Prod. Rep. 29 (2012) 11. [6] M. Potier, L. Mameli, M. Bélisle, L. Dallaire, S.B. Melanc¸on, Anal. Biochem. 94 (1979) 287. [7] J. Kongkamnerd, L. Cappelletti, A. Prandi, P. Seneci, T. Rungrotmongkol, N. Jongaroonngamsang, P. Rojsitthisak, V. Frecer, A. Milani, G. Cattoli, Bioorg. Med. Chem. 20 (2012) 2152. [8] R.C. Buxton, B. Edwards, R.R. Juo, J.C. Voyta, M. Tisdale, R.C. Bethell, Anal. Biochem. 280 (2000) 291. [9] Y.B. Ryu, H.J. Jeong, S.Y. Yoon, J.Y. Park, Y.M. Kim, S.J. Park, M.C. Rho, S.J. Kim, W.S. Leet, J. Agric. Food Chem. 59 (2011) 6467. [10] J. Kongkamnerd, A. Milani, G. Cattoli, C. Terregino, I. Capua, L. Beneduce, A. Gallotta, P. Pengo, G. Fassina, O. Monthakantirat, J. Biomol. Screen. 16 (2011) 755. [11] B.M. Marathe, V. Leveque, K. Klumpp, R.G. Webster, E.A. Govorkova, PLoS ONE 8 (2013) e56704, http://dx.doi.org/10.1371/journal.pone.0056704.
145
[12] S. Liu, J. Yan, J.P. Xing, F.R. Song, Z.Q. Liu, S.Y. Liu, J. Pharmaceut. Biomed. 59 (2012) 96. [13] K. Swaminathan, J.C. Dyason, A. Maggioni, M. von Itzstein, K.M. Downard, Anal. Bioanal. Chem. 405 (2013) 6563. [14] Z.H. Nasser, K. Swaminathan, P. Müller, K.M. Downard, Antivir. Res. 100 (2013) 399. [15] K. Swaminathan, K.M. Downard, Anal. Chem. 84 (2012) 3725. [16] T. Ishimoto, K. Jigawa, T.G. Henares, T. Endo, H. Hisamoto, Analyst 138 (2013) 3158. [17] J. Iqbal, C.E. Muller, Analyst 138 (2013) 3104. [18] T. Nemec, Z. Glatz, J. Chromatogr. A 1155 (2007) 206. [19] X. Hai, X. Wang, M. El-Attug, E. Adams, J. Hoogmartens, A. Van Schepdael, Anal. Chem. 83 (2011) 425. [20] C.L. Cardoso, V.V. Lima, A. Zottis, G. Oliva, A.D. Andricopulo, I.W. Wainer, R. Moaddel, Q.B. Cass, J. Chromatogr. A 1120 (2006) 151. [21] C.L. Cardoso, M.C. de Moraes, R.V.C. Guido, G. Oliva, A.D. Andricopulo, I.W. Wainer, Q.B. Cass, Analyst 133 (2008) 93. [22] Q. Yu, X. Zhan, K. Liu, H. Lv, Y. Duan, Anal. Chem. 85 (2013) 4578. [23] J.W. Goodwin, R.S. Harbron, P.A. Reynolds, Colloid. Polym. Sci. 268 (1990) 766. [24] Y. Yang, K. Hamaguchi, J. Biochem. 87 (1980) 1003. [25] Z. Glatz, Electrophoresis 34 (2013) 12. [26] H.J. Yuk, H.W. Ryu, S.H. Jeong, M.J. Curtis-Long, H.J. Kim, Y. Wang, Y.H. Song, K.H. Park, Food Chem. Toxicol. 55 (2013) 144. [27] Y. Wang, M.J. Curtis-Long, H.J. Yuk, D.W. Kim, X.F. Tan, K.H. Park, Bioorg. Med. Chem. 21 (2013) 6398. [28] R.W. Myers, R.T. Lee, Y.C. Lee, G.H. Thomas, L.W. Reynolds, Y. Uchida, Anal. Biochem. 101 (1980) 166. [29] J.J. Thomas, E.C. Folger, D.L. Nist, B.J. Thomas, R.H. Jones, Anal. Biochem. 88 (1978) 461. [30] A.L. Liu, H.D. Wang, S.M. Lee, Y.T. Wang, G.H. Du, Bioorg. Med. Chem. 16 (2008) 7141. [31] H. Nehmé, R. Nehmé, P. Lafite, S. Routier, P. Morin, J. Chromatogr. A 1314 (2013) 298. [32] B.T. Burlingham, T.S. Widlanski, J. Chem. Educ. 80 (2003) 214.
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Screening of neuraminidase inhibitors from traditional Chinese medicines by integrating capillary electrophoresis with immobilized enzyme microreactor Haiyan Zhao, Zilin Chen ∗ Key Laboratory of Combinatorial Biosynthesis and Drug Discovery (Wuhan University), Ministry of Education, and Wuhan University School of Pharmaceutical Sciences, Wuhan 430071, China
a r t i c l e
i n f o
Article history: Received 30 January 2014 Received in revised form 5 March 2014 Accepted 10 March 2014 Available online 18 March 2014 Keywords: Neuraminidase Inhibitor screening Immobilized enzyme microreactor Short-end injection Traditional Chinese medicine Capillary electrophoresis
a b s t r a c t A simple and effective neuraminidase-immobilized capillary microreactor was fabricated by glutaraldehyde cross-linking technology for screening the neuraminidase inhibitors from traditional Chinese medicines. The substrate and product were separated by CE in short-end injection mode within 2 min. Dual-wavelength ultraviolet detection was employed to eliminate the interference from the screened compounds. The parameters relating to the separation efficiency and the activity of immobilized neuraminidase were systematically evaluated. The activity of the immobilized neuraminidase remained 90% after 30 days storage at 4 ◦ C. The immobilized NA microreactor could be continuously used for more than 200 runs. The Michaelis–Menten constant of neuraminidase was determined by the microreactor as 136.6 ± 10.8 M. In addition, six in eighteen natural products were found as potent inhibitors and the inhibition potentials were ranked in the following order: bavachinin > bavachin > baicalein > baicalin > chrysin and vitexin. The half-maximal inhibitory concentrations were 59.52 ± 4.12, 65.28 ± 1.07, 44.79 ± 1.21 and 31.62 ± 2.04 for baicalein, baicalin, bavachin and bavachinin, respectively. The results demonstrated that the neuraminidase-immobilized capillary microreactor was a very effective tool for screening neuraminidase inhibitors from traditional Chinese medicines. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Influenza virus infection is a major public health problem worldwide, causing serious respiratory illnesses and high death toll each year. Neuraminidase (EC 3.2.1.18, NA) cleaves the specific linkage of the sialic acid receptor to facilitate the release of influenza virus to infect new cells [1]. Additionally, NA may promote the early process of influenza virus infection of lung epithelial cells [2]. Due to the key role in the replication, spread, infection and pathogenesis of influenza virus, the application of neuraminidase inhibitors (NAIs) has been considered as a dominant approach for the treatment of influenza infection. In particular, NAIs are effective for influenza type A and B viruses in human, avian and animal. Among the available neuraminidase inhibitors, oseltamivir and zanamivir are the most widely used drugs. However, their applications are limited due to the rapid emergence of increasing numbers of resistant
∗ Corresponding author at: School of Pharmaceutical Sciences, Wuhan University, Wuhan 430071, China. Tel.: +86 27 68759893; fax: +86 27 68759850. E-mail address: [email protected] (Z. Chen). http://dx.doi.org/10.1016/j.chroma.2014.03.028 0021-9673/© 2014 Elsevier B.V. All rights reserved.
strains during their clinical treatment [3,4], therefore it is necessary to develop novel and effective antiviral agents. A variety of methods have been developed for the discovery of novel NAIs, including NA enzyme-based inhibitory assays, cell culture-based screening, X-ray crystal structure-based computational docking studies and virtual screening [5]. The enzyme-based inhibition assay is the most widely used method by either fluorescence (FL) [6,7] or chemiluminescence (CL) detection [8,9]. However, CL has some disadvantages. For instance, color quenching and cross talk may lead to significant errors. In FL detection, the fluorogenic substrate may nonlinearly interfere with the fluorescence measurement of its product 4-methylumbelliferone (4-MU). The autofluorescence substances such as flavonoids have been reported to exert a considerable quenching effect causing falsepositive results [10]. To solve these problems, a correction factor was introduced according to the fluorescence measurements of 4-MU and 4-MU mixed with substrate [11]. Recently, simple and sensitive affinity screening methods based on MS have been applied to screen potential NAIs in the absence of substrate. Liu et al. developed an HPLC-electrospray ionization tandem mass spectrometric method combined with ultrafiltration for studying NA inhibitory
140
H. Zhao, Z. Chen / J. Chromatogr. A 1340 (2014) 139–145
activity of n-butanol extract of Compound Indigowoad Root Granule [12]. Matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS) has been used to identify the active site domains within influenza NA and to screen inhibitors by control measurements of peptide residues of NA alone and NA-inhibitor complexes after tryptic digestion [13–15]. Among the reported methods, the enzymic reaction was carried out by off-line incubation for 0.5–1 h, which was time-consuming and sample-consuming. As an alternative, online incubation with immobilized enzyme attracts much interest. NA and the substrate were immobilized onto a combinable sensor array for inhibitors assay [16]. The sensor simplified the conventional multiple steps into a single-step operation. However, the method suffers from the enzymic deactivation in three days and the interference from the fluorogenic substrate. The immobilized enzyme microreactor (IMER) has many remarkable advantages such as automatic operation, compatibility with detectors, low consumption of samples, high enzyme activity and reusability. The development and use of IMERs have recently been reviewed [17]. This work aims to develop a new NA-immobilized capillary microreactor integrated with capillary electrophoresis to screen potential NA inhibitors from traditional Chinese medicines. Capillary electrophoresis was employed to separate the substrate and product with diode array detection (DAD). The interference from the screened compounds was eliminated by dual-wavelength detection. To enhance the screening efficiency, a short-end injection procedure [18,19] (sample injected from the capillary outlet end) was introduced by applying a negative pressure at the inlet. The parameters relating to the separation efficiency and NA activity were optimized and the kinetics constant was tested under the optimal experimental conditions. Finally, this method was further verified by eighteen compounds from traditional Chinese medicines. 2. Materials and methods 2.1. Chemicals The substrate 2 -(4-methylumbelliferyl)-␣-d-N-acetylneuraminic acid sodium salt hydrate (4-MuNeu5Ac), the product 4-methylumbelliferone (4-MU) and neuraminidase (10 UN in package, from Clostridium perfringens, Type V, lyophilized powder) were purchased from Sigma–Aldrich (Steinheim, Germany). 3-Aminopropyltriethoxysilane (APTES) was obtained from Aladdin Reagent Co., Ltd. (Shanghai, China). Glutaraldehyde solution in water (25%, m/v) (GA) and some analytical reagents such as sodium dihydrogen phosphate, disodium hydrogen phosphate, sodium acetate anhydrous and acetic acid were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All the screened compounds were obtained from Shanghai Shunbo Bio-engineering Technology (Shanghai, China). Deionized water was purified using a Milli-Q System (Millipore, Bedford, MA, USA). Other reagents used in the experiment were of analytical grade and from commercial sources. 2.2. Instrumentation The capillary electrophoresis system was an Agilent 7100 CE system (Waldbronn, Germany) equipped with a photodiode array detector. CE separations were performed in fused-silica capillary (35.5 cm total length, 8.5 cm from the outlet end to the detection window, 50 m i.d., 365 m o.d.) obtained from Ruifeng Chromatographic Devices (Yongnian, Hebei, China). New capillaries were preconditioned by 1.0 M NaOH for 2 h and deionized water for 0.5 h. After injection into the capillary outlet end by applying a negative
pressure of −5 kPa for 6 s at the inlet, the analytes were separated at +20 kV with a BGE of 10 mM sodium phosphate buffer (pH 5.5). For inhibition assay, baicalein, baicalin and vitexin were detected at 320 nm with a reference wavelength of 260 nm. The other compounds were directly detected at 320 nm with a bandwidth 4 nm. The capillary was thermostatted at 30 ◦ C. Between runs, the capillary was flushed with water and BGE at 95 kPa for 2 min and 5 min, respectively.
2.3. Solution preparation The BGE was prepared by adding 10 mM disodium hydrogen phosphate to 10 mM sodium dihydrogen phosphate under continuous stirring till pH value reached 5.5. Stock solutions of substrate 4-MuNeu5Ac and NA were prepared in deionized water at the concentration of 2 mM and 10 U/mL, respectively. Then the stock solutions were divided into small quantities and stored at −20 ◦ C. Product 4-MU and the screened compounds were dissolved in methanol and stored at 4 ◦ C. The substrate, NA and the screened compounds were separately diluted with pH 5.5 phosphate buffer prior to use each day. All the prepared solutions were filtered by 0.45 m nylon membrane.
2.4. Preparation of immobilized NA capillary microreactor NA was covalently immobilized onto the capillary wall as the previously reported method [20–22] with some modifications. The preparation protocol is shown in Fig. 1. Briefly, the inner surface of the capillary was first coated with APTES by pumping the solution of 30% (v/v) APTES in methanol through the capillary for 5 min. Two ends of the capillary were connected by a piece of Teflon tube and polymerized at 95 ◦ C for 1 h. The capillary was washed with methanol and dried by nitrogen stream. Next, a glutaraldehyde solution 2.5% (v/v) in pH 7.0 sodium phosphate buffer was introduced into the capillary outlet end by capillary wetting to make a 5 cm length coating in the capillary outlet end. Two ends of the capillary were connected by a piece of Teflon tube and reacted in 40 ◦ C water bath for 2 h. Thus the amine groups of APTES bonded to aldehyde group on one side of GA. The capillary was then backflushed with pH 7.0 sodium phosphate buffer for 30 min to remove redundant GA. After dried by nitrogen, the capillary was treated with 0.5 U/mL NA in pH 7.0 sodium phosphate buffer and then incubated at 25 ◦ C for 30 min. Unreacted enzyme was back-flushed out with phosphate buffer, then the capillary was stored at 4 ◦ C for 24 h. Thus, NA was covalently immobilized onto the capillary outlet end by the coupling of amino groups of NA and aldehyde groups on the other side of GA. Before use, the extra 4 cm length of the capillary outlet was cut off to leave 1 cm in length of the immobilized enzyme microreactor. While not in use, the capillary was maintained in pH 7.0 sodium phosphate buffer and stored at 4 ◦ C.
2.5. Kinetics study of NA 1-cm length of substrate plug was injected into the capillary from the outlet end and incubated for 1 min. The substrate and product were separated at +20 kV with BGE of 10 mM sodium phosphate buffer (pH 5.5). Following the on-line enzymic reaction, the rate of hydrolysis of the substrate was calculated according to the increase in the peak area of 4-MU with increasing substrate concentrations at 15, 31, 62.5, 125, 250, 500 and 1000 M. Each concentration was analyzed three times. The data were fitted to double reciprocal plot by Origin Pro 8.0 software (OriginLab, Northampton, MA, USA).
H. Zhao, Z. Chen / J. Chromatogr. A 1340 (2014) 139–145
141
Fig. 1. Procedure for immobilizing NA on capillary inner wall (a) and schematic of on-line enzyme assay (b).
2.6. Inhibitors screening A mixture of substrate and the test compound was diluted with BGE to the final concentration of 125 M and 60 M, respectively. The mixture was injected into the enzyme microreactor at −5 kPa for 6 s. After incubation for 4 min, the product was separated from the unreacted substrate under the optimal conditions, and each inhibitor assay was performed in triplicate (n = 3). Before each analysis, the capillary was flushed with water for 2 min and conditioned with BGE for 5 min. The inhibitors were identified by the reduction of the peak area of the product. The percentages of inhibition were calculated by the equation: inhibition% =
1−
AI AS
× 100%
(1)
at pH < 6.0 (isoelectric point of APTES at pH 6.0) [23], the surface of the modified capillary inner wall was positively charged, which resulted in a reversed EOF from the cathode to the anode. Hence, the separation voltage was set at +20 kV at pH of 3.0–5.5 and −20 kV at pH 6.0 for the injection from the outlet end. As shown in Fig. 2, the peak area of 4-MU and the resolution increased as the increase of pH. The analytes could be well separated at the all tested pH. The migration time of analytes was much dependent on the pH of BGE. The lower the BGE pH, the shorter was the migration time. This was due to the fact that the protonations of amino groups in APTES at lower pH would give rise to higher effective mobility and faster electrophoretic migration of the analytes. Because of its pKa value 7.8 [24], 4-MU is electric neutral and migrated with the same velocity as the EOF at the investigated pH range. Since the substrate was negatively charged at pH < 6.0, it electrophoretically migrated
where AI and AS represent the peak area of 4-MU obtained in the presence and absence of inhibitor, respectively. Half-inhibitory concentration (IC50 ) was determined at constant substrate concentration of 125 M with eight concentrations of inhibitors ranged from 2 to 320 M. Each concentration of inhibitor was assayed in triplicate. The dose-response curve for enzyme inhibition was plotted by fitting the percentage of enzyme activity as a function of log [inhibitor]. The enzyme activity (%) was defined as the ratio of the peak area of 4-MU in the presence and absence of the inhibitor. IC50 was defined as the inhibitor concentration necessary to reduce the enzyme activity by 50%. The data processing and curve-fitting were performed by PRISM® 5.0 (GraphPad software, San Diego, CA). 3. Results and discussion 3.1. Optimization of CE separation conditions The substrate and product were separated by capillary electrophoresis in fused silica capillary with modification of APTES. The BGE pH was optimized on the basis of the effect of pH (from 3.0 to 6.0) on the migration time and peak area of 4-MU (Fig. 2). Since the primary amino groups of APTES were protonated to be cationic
Fig. 2. Effect of BGE pH on the separation efficiency of substrate and product. CE conditions: fused-silica capillary with modification of APTES, 50 m i.d. × 35.5 cm (8.5 cm from the outlet end to the detector); BGE, 10 mM sodium phosphate buffer; separation voltage, +20 kV for pH 3.0–5.5 and −20 kV for pH 6.0; sample injection, −5 kPa, 6 s; UV detection at 320 nm; cartridge temperature, 30 ◦ C; concentration of substrate and product, 125 and 200 M, respectively.
142
H. Zhao, Z. Chen / J. Chromatogr. A 1340 (2014) 139–145
Fig. 3. Effect of incubation time (a and b), pH of sample solution (c), and the percentage of methanol (d) on the peak area of 4-MU by on-line enzymic reaction. CE conditions: fused-silica capillary with immobilized enzyme microreactor, 50 m i.d. × 35.5 cm (8.5 cm from the outlet end to the detector); sample injection, −5 kPa, 6 s; incubation time, 4 min; BGE, 10 mM pH 5.5 sodium phosphate buffer; separation voltage, +20 kV; UV detection at 320 nm; cartridge temperature, 30 ◦ C; concentration of substrate, 125 M.
in the same direction as EOF and faster than the product. At pH 6.0, the substrate electrophoretically migrated in the reverse direction of EOF and faster than EOF, therefore it would fall back into the outlet vial. To achieve shorter migration time and higher peak area of 4-MU, pH 5.5 was chosen as the optimum pH of BGE. To shorten the analytical time, the routine injection from the capillary inlet end (long-end injection mode) was replaced by the injection from the outlet end of the capillary (short-end injection mode). It has been demonstrated that outlet injection can greatly reduce the analysis time and enhance the sensitivity of detection [25]. In this study, the capillary effective length was shortened to 8.5 cm and the separation current was maintained. As shown in Fig. S1, the migration time of the product in long-end injection mode was four times longer than that in short-end injection mode.
3.2. Optimization of online immobilized NA reaction conditions It is necessary for the substrate to online incubate with the immobilized NA for several minutes to generate adequate product. The incubation time varied from 50 s to 600 s was used to examine the effect of incubation time on the peak area of 4-MU. The peak area of 4-MU increased linearly over the first 4 min with a determination coefficient of 0.9949 (Fig. 3a–b) but increased slightly after 4 min, which indicated that the incubation for 4 min was sufficient for the enzymic reaction. Thus, the incubation for 4 min was chosen to carry out the further experiments. To investigate the pH effect of sample solution on the reaction efficiency, the substrate was diluted by buffers with different pH values. The peak areas of the product showed no significant difference in pH span 3.0–7.0 whereas dramatically decreased at pH 8.0 (Fig. 3c), which indicated that pH 3.0–7.0 was feasible for additional enzymic reaction. In many investigations, pH 4.2–5.0 was commonly used for the enzymic reaction [26,27]. Additionally, the enzyme immobilization failed at pH value lower than 5.0 in this
study. Taking overall consideration of the IMER stability, the BGE for separation and the immobilized enzyme activity, pH 5.5 was selected as the optimal reaction condition. The effect of methanol on the reaction efficiency was evaluated by the substrate dilution with buffer containing different percentages of methanol. The peak area of 4-MU decreased slightly with the increase of methanol from 0 to 30% but significantly with the increase of methanol higher than 30% (v/v) (Fig. 3d). To avoid false results in inhibition assay, the final methanol concentration in the sample should be lower than 10% (v/v). 3.3. Method validation The reproducibility of the immobilization procedure was assessed by measuring the peak areas of 4-MU from enzymic reactions of six similar microreactors at fixed substrate concentration of 125 M. Each microreactor was assayed in five times. The RSD for the migration time and peak area of the product were 3.71% and 5.33%, respectively. The storage stability was examined by measuring the peak area of the product after storage at 4 ◦ C in BGE for 30 days, and the activity of immobilized NA retained 90% with RSD of 3.81%. Furthermore, the immobilized NA microreactor could be continuously used for at least 200 runs. The results showed that the microreactor had good stability. The intra-day and inter-day precision was estimated by five runs of on-line enzymic reaction at fixed substrate concentration of 125 M on the same day and on five consecutive days, respectively. The results showed that the RSD of the peak area of the product was 3.33% for intra-day precision and 6.15% for inter-day precision. 3.4. Kinetics study of immobilized NA For enzyme-catalyzed reactions, the relation between initial reaction velocity (V0 ) and substrate concentration [S] can be
H. Zhao, Z. Chen / J. Chromatogr. A 1340 (2014) 139–145
143
Fig. 4. Michaelis–Menten plot (a) and double reciprocal plot (b) at varied concentrations of substrate ranging from 15 to 1000 M by online enzymic reaction. Incubation time, 1 min; other CE conditions as in Fig. 3.
described by the Michaelis–Menten equation, and the kinetics parameters Km and Vmax can be calculated by linear regression from Lineweaver–Burk equation: Km 1 1 1 = + Vmax [S] Vmax V0
(2)
where Vmax is the maximum reaction velocity and Km is the Michaelis–Menten constant, the substrate concentration at half of the maximum velocity. To determine the kinetic constant (Km ), the conversion rate of the substrate should be less than 10%, hence incubation time was set at 1 min. The peak area of the product was used to indicate the initial reaction velocity (V0 ). After the online enzymic reaction, the Km was estimated by linear regression from double reciprocal plots with substrate concentrations ranging from 15 to 1000 M (Fig. 4). Each concentration was analyzed in triplicate (n = 3). The linear regression equation of the double reciprocal curve was y = 15.68x + 0.1148 with a determination coefficient of 0.9925. From the slope and the intercept on the ordinate, the Km value for the immobilized NA was calculated to be 136.6 ± 10.8 M which was in the range of 102–150 M reported in literatures [6,28,29]. 3.5. Inhibition assay by online NA-immobilized capillary microreactor Initially, the substrate, 4-MU and the screened compounds were separated by capillary electrophoresis under the optimal conditions. Baicalein and vitexin had the same migration time as 4-MU and baicalin had the same migration time as the substrate (Fig. S2a), which suggested that the three compounds may interfere with the detection of substrate and 4-MU. Dual-wavelength detection was adopted to eliminate this interference by setting the equal absorbance wavelength of baicalein, baicalin and vitexin as the reference wavelength in DAD. Since the absorbances of the three compounds at 320 nm were equal to those at 260 nm whereas 4MU had scarce absorption at 260 nm (Fig. S3), 260 nm was chosen as the reference wavelength (see the supporting information). As shown in Fig. S2b, the peaks of the three compounds disappeared, while the peak height of the product remained almost unchanged. Under the optimal experimental conditions, the NAimmobilized microreactor was applied to investigate the inhibition potential of eighteen natural products. The stock solutions of these compounds were individually mixed with the substrate stock solution and diluted by the BGE to achieve the final concentration of the substrate at 125 M and the screened compounds at 60 M. These compounds were kept at constant concentration for ranking
Table 1 Inhibitors screening by immobilized neuraminidase microreactor (n = 3). Compounds
I%
Baicalein Baicalin Vitexin Chrysin Bavachin Bavachinin Resveratrol Psoralen Isopsoralen
62.4 ± 56.7 ± 42.3 ± 43.5 ± 81.1 ± 85.1 ± 0% 0% 0%
5.1% 3.5% 3.5% 2.3% 7.8% 6.4%
Compounds
I%
Oleanolic acid Ursolic acid Demethoxycurcumin Lycorine Galanthamine Matrine Oxymatrine Aconitine Vinblastine
0% 0% 0% 0% 0% 0% 0% 0% 0%
their inhibition potential. The overlaid electropherograms of online enzymic reaction in the presence of inhibitors are shown in Fig. 5 and the percentages of inhibition are summarized in Table 1. The results indicated that six natural products showed potential inhibition for NA. The inhibition potentials were ranked in the following order: bavachinin > bavachin > baicalein > baicalin > chrysin and vitexin. The ranking order of baicalein, baicalin, chrysin and vitexin was consistent with that in the literature [30]. Structure–activity study demonstrated that the inhibition activity would reduce when hydroxyl groups on C-5 or C-7 position were replaced by glycosyl groups, whereas remained unchanged when replaced by methoxyl groups. Due to the lack of C4 -OH, chrysin and vitexin displayed weaker activity [30]. The inhibition effects of bavachinin and bavachin were firstly investigated in our work. The dose-response curves for four potent inhibitors showed that these compounds had dose-dependent inhibitory effects against immobilized NA (Fig. 6). The enzyme activity decreased with the increasing of the inhibitor concentrations, which indicated that the enzyme activity was inhibited in the capillary after online enzymic reaction. The IC50 values were determined as 59.52 ± 4.12, 65.28 ± 1.07, 44.79 ± 1.21 and 31.62 ± 2.04 for baicalein, baicalin, bavachin and bavachinin, respectively. Due to their poor solubility, vitexin and chrysin were dissolved in high percentage of methanol which would lead to false-positive results. Therefore, the IC50 values of vitexin and chrysin were not determined. The IC50 values of baicalin and baicalin were higher than those reported in the literature [12]. The difference can be explained by the fact that the concentrations of the inhibitors increased after ultrafiltration [12]. The magnitude of IC50 depends upon the concentration of the substrate, origin of the enzyme and the assay conditions. Hence, the absolute value of IC50 is less important than the order of magnitude and the relative value between two inhibitors [31,32].
144
H. Zhao, Z. Chen / J. Chromatogr. A 1340 (2014) 139–145
Fig. 5. The overlaid electropherograms for inhibitors screening by online immobilized enzymic assay detected at 320 nm with a reference wavelength of 260 nm (a) and at 320 nm without reference wavelength (b). CE conditions as in Fig. 3.
Fig. 6. Dose-response curves fit of neuraminidase activity (%) as a function of log [inhibitor]. Each point represents average activity and error bars indicate the SD; n = 3.
4. Conclusions The first online NA inhibitor screening method was successfully established by CE based on immobilized NA capillary microreactor. The method was effective and low cost because of the reusable enzyme and the nanoliter level consumption of the samples. The integration of enzymic reaction and separation into one capillary rendered the inhibitors screening fully automated. The short-end injection mode greatly enhanced the screening efficiency. Additionally, the dual-wavelength detection eliminated the interference of the screened compounds and minimized the
false-positive results. The inhibitors screening method based on immobilized NA microreactor developed in this study provided a framework for discovering new NAIs.
Acknowledgements This work was supported by the National Natural Science Foundation of China (Grant nos. 21375101 and 30973672), Wuhan Science and Technology Bureau (No. 20140601010057), Doctoral Fund of Ministry of Education of China (No. 20110141110024),
H. Zhao, Z. Chen / J. Chromatogr. A 1340 (2014) 139–145
Innovation Seed Fund and Translational Medical Research Fund of Wuhan University School of Medicine. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.chroma.2014. 03.028. References [1] G. Neumann, T. Noda, Y. Kawaoka, Nature 459 (2009) 931. [2] M.N. Matrosovich, T.Y. Matrosovich, T. Gray, N.A. Roberts, H.D. Klenk, J. Virol. 78 (2004) 12665. [3] A.J. Hay, F.G. Hayden, Lancet 381 (2013) 2230. [4] Y. Matsuzaki, K. Mizuta, Y. Aoki, A. Suto, C. Abiko, K. Sanjoh, K. Sugawara, E. Takashita, T. Itagaki, Y. Katsushima, Virol. J. 7 (2010) 53. [5] U. Grienke, M. Schmidtke, S. von Grafenstein, J. Kirchmair, K.R. Liedl, J.M. Rollinger, Nat. Prod. Rep. 29 (2012) 11. [6] M. Potier, L. Mameli, M. Bélisle, L. Dallaire, S.B. Melanc¸on, Anal. Biochem. 94 (1979) 287. [7] J. Kongkamnerd, L. Cappelletti, A. Prandi, P. Seneci, T. Rungrotmongkol, N. Jongaroonngamsang, P. Rojsitthisak, V. Frecer, A. Milani, G. Cattoli, Bioorg. Med. Chem. 20 (2012) 2152. [8] R.C. Buxton, B. Edwards, R.R. Juo, J.C. Voyta, M. Tisdale, R.C. Bethell, Anal. Biochem. 280 (2000) 291. [9] Y.B. Ryu, H.J. Jeong, S.Y. Yoon, J.Y. Park, Y.M. Kim, S.J. Park, M.C. Rho, S.J. Kim, W.S. Leet, J. Agric. Food Chem. 59 (2011) 6467. [10] J. Kongkamnerd, A. Milani, G. Cattoli, C. Terregino, I. Capua, L. Beneduce, A. Gallotta, P. Pengo, G. Fassina, O. Monthakantirat, J. Biomol. Screen. 16 (2011) 755. [11] B.M. Marathe, V. Leveque, K. Klumpp, R.G. Webster, E.A. Govorkova, PLoS ONE 8 (2013) e56704, http://dx.doi.org/10.1371/journal.pone.0056704.
145
[12] S. Liu, J. Yan, J.P. Xing, F.R. Song, Z.Q. Liu, S.Y. Liu, J. Pharmaceut. Biomed. 59 (2012) 96. [13] K. Swaminathan, J.C. Dyason, A. Maggioni, M. von Itzstein, K.M. Downard, Anal. Bioanal. Chem. 405 (2013) 6563. [14] Z.H. Nasser, K. Swaminathan, P. Müller, K.M. Downard, Antivir. Res. 100 (2013) 399. [15] K. Swaminathan, K.M. Downard, Anal. Chem. 84 (2012) 3725. [16] T. Ishimoto, K. Jigawa, T.G. Henares, T. Endo, H. Hisamoto, Analyst 138 (2013) 3158. [17] J. Iqbal, C.E. Muller, Analyst 138 (2013) 3104. [18] T. Nemec, Z. Glatz, J. Chromatogr. A 1155 (2007) 206. [19] X. Hai, X. Wang, M. El-Attug, E. Adams, J. Hoogmartens, A. Van Schepdael, Anal. Chem. 83 (2011) 425. [20] C.L. Cardoso, V.V. Lima, A. Zottis, G. Oliva, A.D. Andricopulo, I.W. Wainer, R. Moaddel, Q.B. Cass, J. Chromatogr. A 1120 (2006) 151. [21] C.L. Cardoso, M.C. de Moraes, R.V.C. Guido, G. Oliva, A.D. Andricopulo, I.W. Wainer, Q.B. Cass, Analyst 133 (2008) 93. [22] Q. Yu, X. Zhan, K. Liu, H. Lv, Y. Duan, Anal. Chem. 85 (2013) 4578. [23] J.W. Goodwin, R.S. Harbron, P.A. Reynolds, Colloid. Polym. Sci. 268 (1990) 766. [24] Y. Yang, K. Hamaguchi, J. Biochem. 87 (1980) 1003. [25] Z. Glatz, Electrophoresis 34 (2013) 12. [26] H.J. Yuk, H.W. Ryu, S.H. Jeong, M.J. Curtis-Long, H.J. Kim, Y. Wang, Y.H. Song, K.H. Park, Food Chem. Toxicol. 55 (2013) 144. [27] Y. Wang, M.J. Curtis-Long, H.J. Yuk, D.W. Kim, X.F. Tan, K.H. Park, Bioorg. Med. Chem. 21 (2013) 6398. [28] R.W. Myers, R.T. Lee, Y.C. Lee, G.H. Thomas, L.W. Reynolds, Y. Uchida, Anal. Biochem. 101 (1980) 166. [29] J.J. Thomas, E.C. Folger, D.L. Nist, B.J. Thomas, R.H. Jones, Anal. Biochem. 88 (1978) 461. [30] A.L. Liu, H.D. Wang, S.M. Lee, Y.T. Wang, G.H. Du, Bioorg. Med. Chem. 16 (2008) 7141. [31] H. Nehmé, R. Nehmé, P. Lafite, S. Routier, P. Morin, J. Chromatogr. A 1314 (2013) 298. [32] B.T. Burlingham, T.S. Widlanski, J. Chem. Educ. 80 (2003) 214.
