Synthesis and Adsorption Performance of Molecularly Imprinted Microspheres for Quercetin by Suspension Polymerization

Quercetin (3,3’,4,5,7-penta-hydroxy flavone) is a plant-derived flavonoid found widely occurring in leaves, fruits, vegetables and grains [1]. Quercetin has become a hot topic based on its various bioactivities such as anti-inflammation, inoxidability, antiviral, antitumor and innate immune function [2]. The analysis of quercetin involve in HPLC-UV [3-5], electrogenerated chemiluminescence [6] and capillary electrophoresis [7]. Because of low concentrations of quercetins in nature, the complexity of samples and the structural similarity to other flavonols, selective extract methods are necessary prior to analysis.


Introduction
Quercetin (3,3',4,5,7-penta-hydroxy flavone) is a plant-derived flavonoid found widely occurring in leaves, fruits, vegetables and grains [1]. Quercetin has become a hot topic based on its various bioactivities such as anti-inflammation, inoxidability, antiviral, antitumor and innate immune function [2]. The analysis of quercetin involve in HPLC-UV [3][4][5], electrogenerated chemiluminescence [6] and capillary electrophoresis [7]. Because of low concentrations of quercetins in nature, the complexity of samples and the structural similarity to other flavonols, selective extract methods are necessary prior to analysis.
Some studies about determination and separation for quercetin using MIPs have been reported [4,[20][21][22]. However, these quercetin-MIPs were prepared by bulk polymerization, and the required particles were obtained with crushing, grinding and sieving. As well as we known, the bulk polymerization method has many limitations, such as time-consuming, waste, homogeneous absorption sites, poor reproducibility and selectivity [23].
In this work, the preparation of molecularly imprinted microspheres for quercetin by aqueous suspension polymerization was first presented. The imprinting effect of the imprinted microspheres was evaluated, and selectively recognition capability was demonstrated. The equilibrium binding experiment shows that the maximum theoretical static binding capacity of the imprinted microspheres for quercetin can be up to 96.5927 mg g -1 .

Preparation of molecularly imprinted microspheres for quercetin
Molecularly imprinted microspheres for quercetin were prepared by aqueous suspension polymerization methods as follows: quercetin, MAA, EGDMA and AIBN were dissolved in porogen solvent ethyl acetate (also used as oil phase). Then it was purged with high purity nitrogen for 20 min, and added into 1.5% wt. PVA/water in a glass vessel and stirred at 350 rpm. Polymerization was carried out at 60°C for 24 h under stirring and N 2 atmosphere.
The resulted microspheres were washed by hot water to remove PVA. Then they were washed respectively by 15% acetic acid-methanol and methanol/water solution under ultrasonic until no quercetin was detected in the rinse solution.
Non-imprinted polymers (NIPs) for control experiments were obtained following the procedure described above but excluding quercetin from the formulation. microspheres were dried at 60°C and then attached to the silver paper and coated with a gold layer. A scanning electron microscope (JSM-6700, JEOL SEM Co., Japan) was used for the morphology observation.

HPLC-UV analysis of analytes
The HPLC-UV analysis was carried out using a Shimadzu-LC-20AT pump equipped with SIL-20A injector, a Shimadzu SPD-M20A US detector and a CTO-2AScolumn oven. Column oven temperature is controlled at 25°C. The column is C 18 column (4.6mm × 250mm, i.d.). Analytical procedure was conducted according to the reference [22]. That is, the column was equilibrated by mobile phase that acetonitrile and 0.1% (w/v) H 3 PO 4 aqueous solution (36: 64, v/v). The detection was run with wavelength at 365 nm. The temperature was controlled at 25°C. The flow rate was maintained at 0.6 mL min −1 and 20 µL of sample was injected.

Binding analysis of molecularly imprinted microspheres
The imprinted microspheres or non-imprinted microspheres of 25 mg were mixed with 5 mL of solution containing a known concentration of quercetin in methanol/water (95: 5, v/v). The mixture was stirred at 150 rpm and incubated for 3 h at room temperature. The concentration of the free substrate in the supernatant solution was determined by HPLC-UV analysis after the sample was centrifuged at 6000 rpm for 12 min. The binding capacity (Q) was defined as mg of substrate bound per 1 g microspheres, and calculated by the change of quercetin concentration after and before adsorption by equation (1), in which C 0 (mg mL -1 ) and C (mg mL -1 ) are the initial concentration and free concentration of substrate in the supernatant, respectively. V (mL) is the volume of adsorption solution. W(g) is the mass of the microspheres.

Selectivity of the imprinted microspheres
Three kinds of flavones (quercetin, isorhamnetin, rutin) were selected as competitive agents to estimate selectivity of MIPs for quercetin. Structures of quercetin, isorhamnetin and rutin were shown in Figure 1. MIP/NIP microspheres (25 mg) were respectively dispersed in 6 vials containing 5 mL of methanol/water (95: 5, v/v) solution of the substrate. And initial concentrations of quercetin, isorhamnetin and rutin are 80 mg L -1 . The mixture is stirred at 150 rpm and incubated for 3 h at room temperature, and equilibrium concentrations of quercetin, isorhamnetin and rutin are determined by HPLC-UV analysis.

Preparation and evaluation of molecularly imprinted microspheres
There are a few reports about the preparation of MIPs for quercetin, but these MIPs were prepared by bulk polymerization followed by grinding to particles. These imprinted particles are polydisperse, irregular and containing a large portion of fine particles. Furthermore, extensive sieving and sediment are necessary in order to achieve a rather narrower size distribution and remove fine particles. They are time-consuming and labor-wasting, and the imprinting effect is always not satisfactory [23].
In this work, the imprinted microspheres for quercetin were prepared by aqueous suspension polymerization, in which water solution containing 1.5% PVA was used as continuous phase, and ethyl acetate solution (it contains quercetin, MAA, EGDMA and AIBN) was used as disperse phase. The ratio of the template molecule quercetin to functional monomer MAA was optimized. The composition of prepolymerization mixtures and imprinting factor (IF) for MIPs and NIPs are shown in Table 1.
These results indicate that when the ratio of quercetin to MAA is 1:4, the imprinting effect is the best, and the imprinting factor (the amount of quercetin bound by MIPs / the amount of quercetin bound by NIPs) is the highest. It can be explained as follows: when the ratio of quercetin to MAA during 1:1 ~ 1:3, IF is lower because the insufficient amount of functional monomers in MIPs leads to less binding sites. On the other hand, while the ratio of quercetin to MAA is 1:5 and 1:6, IF is also lower because that the excessive functional monomer MAA in MIPS make non-selective binding sites increase, which creates the enhancement of non-specific interaction between quercetin and MAA. So, the ratio of template quercetin to functional monomer MAA is 1:4 in the following experiments.

Characterization of microspheres
The imprinted microspheres were detected by scanning electron microscope (SEM), and the results were shown in Figure 2. SEM micrographs of the imprinted microspheres reveal that the imprinted microspheres have good sphericity and smooth surface, and there are many micropores in the surface of the microspheres.

Selectivity of the imprinted microspheres
In order to study the recognition capability of the imprinted microspheres, the binding amounts for quercetin, isorhamnetin and rutin by the imprinted microspheres were determined using HPLC-UV analysis. The results are shown in Figure 3, and it indicates that the binding amounts of MIPs for quercetin is much more than that of isorhamnetin and rutin, which further demonstrates the imprinted microspheres have good selective recognition for the template molecule quercetin. The reason is that strong hydrogen bonds are formed between -OH of quercetin and -COOH of MAA in the solvent. In addition, Schematic diagram of the interaction between quercetin and MAA is shown in Figure 4.

Adsorption kinetics of the imprinted microspheres for quercetin
The adsorption kinetics behavior of the imprinted microspheres for quercetin was investigated by changing the adsorption time from 0 min to 180 min. The initial concentration of quercetin was kept a constant of 80 mg L -1 in methanol/water (95: 5, v/v). Adding 25 mg of microspheres into 5 mL solution, then they were mixed at room temperature under stirring at 150 rpm. After being absorbed for some time, each sample was centrifuged and 20 µL of supernatant was assessed by HPLC-UV analysis.
The adsorption dynamic curves of MIPs and NIPs were shown in Figure 5. It can be seen that the amount of quercetin binding to the MIPs is much higher than that of the NIPs. Moreover, to the MIPs, there exists a rapid dynamic adsorption of quercetin to the MIPs. In the first 40 min, the adsorbed amount increases fast with the increase of adsorption time, and then the adsorbed amount keeps a constant in the succedent time. However, there is no large difference among the binding amount of the NIPs for quercetin with the time.

Adsorption isotherm of the imprinted microspheres
It is very important to investigate adsorption isotherm of the imprinted microspheres for further study. The method usually utilized to investigate the thermodynamic adsorption properties of MIPs is to plot an adsorption curve. The experiments were carried out in various quercetin concentrations (1.0×10 -4 ~ 50×10 -4 mol L -1 ) in methanol/ water (95: 5, v/v), in which 25 mg of microspheres (MIP4 and NIP4 in the Table 1) were mixed with 5 mL of quercetin solution for 3 h. Figure 6a shows adsorption isotherms of MIPs and NIPs. The results suggest the binding amounts of MIPs and NIPs for quercetin improve with the increase of concentration of quercetin. Moreover, the former is always much larger than the latter. When initial concentration of quercetin is 30×10 -4 mol L -1 , the adsorbed amount is about 77 mg g -1 toward the MIPs and 23 mg g -1 toward the NIPs, respectively, and the imprinting factor is 3.47, it indicates that the obtained MIPs by our methods have high specific adsorption for the template molecule. It is because that there are apparent differences in tri-dimensional structure between the MIPs and the NIPs. In the MIPs, there are a lot of sites and cavities which are complementary to the template quercetin in size and shape, and they are contributive to high effective selectivity for the template. To NIPs, however, there are no sites and cavities complementary to the template and so their selectivity for quercetin is worse.
In addition, the plotted adsorption isotherm curve of MIPs is shown in Figure 6b. The plotted curve was approximated by modified Langmuir isotherm (equation 2) with parameters a=96.5927, b=0.0173 and c=-0.6340, and the correlative coefficient R 2 = 0.9951. The result

Conclusion
The molecularly imprinted microspheres for quercetin was prepared by using suspension polymerization, in which 1.5% PVA-water solution is used as continuous phase, ethyl acetate solution including quercetin, MAA, EGDMA and AIBN used as disperse phase. The optimal molar ratio of quercetin to MAA is 1:4. The imprinted microspheres were characterized by scanning electron microscopy. Furthermore, the imprinting effect and selectivity of the imprinted microspheres were   The molar ratio of QCT to MAA to EGDMA. investigated, and the results show that the imprinted microspheres can specifically recognize quercetin from other three kinds of antibiotics. The equilibrium binding experiments were employed to study the binding behavior of the imprinted microspheres. The results indicate that the maximum theoretical static binding capacity of molecularly imprinted microspheres can be up to 96.5927 mg g -1 .
Therefore, the resulted imprinted microspheres for quercetin have good selectivity and high binding capacity, and they are promising to be used as adsorption media, even also used as chromatographic stationary phase.