Preparation of phenylboronate affinity rigid monolith with macromolecular porogen

Accepted Manuscript Title: Preparation of phenylboronate affinity rigid monolith with macromolecular porogen Author: Xiang-Jie Li Jia Man Yong-Xin Zhao Zhao-Sheng Liu Haji Akber Aisa PII: DOI: Reference:

S0021-9673(16)30130-3 http://dx.doi.org/doi:10.1016/j.chroma.2016.02.031 CHROMA 357305

To appear in:

Journal of Chromatography A

Received date: Revised date: Accepted date:

4-1-2016 9-2-2016 9-2-2016

Please cite this article as: Xiang-Jie Li, Jia Man, Yong-Xin Zhao, ZhaoSheng Liu, Haji Akber Aisa, Preparation of phenylboronate affinity rigid monolith with macromolecular porogen, Journal of Chromatography A http://dx.doi.org/10.1016/j.chroma.2016.02.031 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Preparation of phenylboronate affinity rigid monolith with macromolecular porogen Xiang-Jie Lia,b,c, Jia Mana,b,c, Yong-Xin Zhaoa, Dr. Zhao-Sheng Liua,b* [email protected], Haji Akber Aisa a,b* [email protected] a

Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Urumqi 830011, Xinjiang, China b State Key Laboratory Basis of Xinjiang Indigenous Medicinal Plants Resource Utilization, Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Urumqi 830011, China c University of Chinese Academy of Sciences, Beijing 100039, China *

Correspondence author at: Xinjiang Technical Institute of Physics and Chemistry,

Chinese Academy of Sciences, Urumqi 830011, Xinjiang, China, Fax: +86-22-23536746.

1

Highlights > Boronate-affinity monolith was first prepared via macromolecular porogen. > The monolithic polymer was made using 4-VPBA, EDMA, and a mixture of PS solution in THF. > The influence of synthesis parameters on affinity was studied. > The Boronate-affinity monolith can extract cis-diol flavonoid glycosides from natural product.

2

Abstract Boronate-affinity monolithic column was first prepared via polystyrene (PS) as porogen

in

this

work.

The

monolithic

polymer

was

synthetized

using

4-vinylphenylboronic acid (4-VPBA) as functional monomer, ethylene glycol dimethacrylate (EDMA) as crosslinker monomer, and a mixture of PS solution in tetrahydrofuran, the linear macromolecular porogen, and toluene as porogen. Isoquercitrin (ISO) and hyperoside (HYP), isomer diol flavonoid glycosides, can be baseline separated on the poly(VPBA-co-EDMA) monolith. The effect of polymerization variables on the selectivity factor, e.g., the ratio of monomer to crosslinker (M/C), the amount of PS and the molecular weight of macromolecular porogen was investigated. The surface properties of the monolithic polymer were characterized by scanning electron microscopy and nitrogen adsorption. The best polymerization condition was the M/C ratio of 7:3, and the PS concentration of 40 mg/ml. The poly(VPBA-co-EDMA) polymer was also applied to extract cis-diol flavonoid glycosides from the crude extraction of cotton flower. After treated by poly(VPBA-co-EDMA) for solid phase extraction, high purity ISO and HYP (>99.96%) can be obtained with recovery of 83.7% and 78.6%, respectively.

Keyword: Monolithic column; phenylboronate affinity monolith; flavonol glycosides; solid phase extraction; macromolecular porogen; HPLC

3

1. Introduction Boronate affinity materials have emerged as important media of molecular recognition and selective separation for cis-diol-containing compounds [1]. The primary mechanism of retention mainly relies on the reversible formation of cyclic, anionic esters, in which cyclic boronate esters are selectively formed with 1,2- and 1,3-cis-vicinal diol moieties at high pH (capturing) and the release of the compounds can be achieved at low pH values. As unique sorbents, boronate affinity materials exhibit several significant advantages, including broad-spectrum selectivity, reversible covalent binding, fast association/desorption kinetics, and good compatibility with mass spectrometry in additional to pH-controlled capture/release. Based on this capture/release principle, numerous studies have been reported on chemical sensing [2], capture of antibodies [3] and affinity chromatography [4,5]. Macroporous monoliths are defined as continuous stationary phases that form as a homogeneous column in a single piece [6]. Compared with conventional chromatographic columns, monolithic columns can provide several significant advantages such as ease of preparation, low cost, low back pressure, and fast mass transfer. Since the boronate affinity monolithic column was first reported in 2006 [7], they have been rapidly developed because of the merits of both boronate affinity and monolithic columns. Especially, boronate affinity monolith in capillary [8-16] has the enhanced chromatographic resolution, higher efficiency, lower sample consumption, more convenient online coupling to mass spectrometry and improved mass-detection sensitivity, as compared with classical column-based HPLC.

4

However, preparing monolithic polymer in wide bore tubes, i.e., polymer rod, is still a challenge. This may be attributed to the consequences of a strong thermal gradient across the wide bore mold in which the monolith is made [6]. Once a solid network is formed during polymerization, further reactions tend to make the monolith shrink since the center of the bed leads to be hotter than the region close to the wall, which causes a mechanical stress at the interface between the monolith and the column wall. If the stress exceeds a certain threshold, the interface breaks and the monolith separates from the wall. This would cause the mobile phase stream to flow between wall and monolith, leading to dramatic loss of retention and efficiency. Whether the rod has shrunk, snapped from the wall and had to be encapsulated (rod columns) or the interface between the monolith and the wall has deformed under stress, depends significanly on the phase separation during the polymeization. In the case of the ordinary polymer monolith made with the porogenic poor solvents, however, the growing polymer chains tend to aggregate each other because van der Waals attraction surmounts the steric hindrance mutually expelling the polymer chains [6]. Thus, the phase separation between growing polymer chains and porogenic solvent proceeds so fast, and the coarsening of monolithic structure inherently leads to heterogeneous macroporous structures composed of tiny micron size globular particles. Therefore, it is desire to suppress the inherent tendency of the aggregated globular monolith structure by searching of appropriate porogenic solvent. Recently, it has been found that the phase separation of a polymer solution has very different internal dynamics from that of classical fluid mixtures [17, 18], in

5

which viscoelastic effects play important roles. This type of phase separation is characterized by a crossover between the characteristic deformation rate induced by phase separation itself and the rheological relaxation rate of the phase rich in the slow component, which can be viewed as viscoelastic relaxation for pattern evolution [19, 20]. For example, in the case where the large difference of molecular dynamics or mobility may exist between solvent molecules and polymeric molecules, the polymer rich phase becomes a transient gel rapidly and this transient gel has its temporal relaxation modulus, i.e., viscoelasticity, resisting the phase separation until the complete relaxation [17]. Thus, the freezing of this transient polymer network structure induced by viscoelastic phase separation tend to expand and may not agglomerate each other, different from those prepared in small molecule porogen. Use of polymer porogens to prepare macroporous beads has been suggested first by Abrams in the early 1950s [21]. In addition, boronate affinity monolith using PEG 10,000 as macromolecular porogen have been reported in capillaries [22,23]. In view of facts above, we intend to propose a new approach for the synthesis of phenylboronate affinity monolith in a format of rod columns. For this purpose, a linear macromolecule, PS, was used as porogenic solvent. In the present study, 4-vinylbenzene boric acid (VPBA) was chosen as functional monomer, and ethylene glycoldimethacrylate (EDMA) as crosslinker since poly(VPBA-EDMA) can capture cis-diol-containing small molecules selectively due to significant hydrophobicity of the crosslinker [1]. In the present work, the boronate affinity monolith made in PS solution was further used for the separation of flavonol glycosides (Fig. 1).

6

2 Experimental 2.1 Materials Hyperoside (HYP, 98%), isoquercitrin (ISO, 98%) and other flavonol glycosides used were purchased from Shifeng Biotechnology Co., Ltd. (Shanghai, China). Trifolin (TRI, 98%) were from Xili Biotechnology Co., Ltd (Yunnan, China). Polystyrene (MW = 150,000, 280,000, 350,000), ethylene glycoldimethacrylate (EDMA, 98%) and 4-vinylbenzene boric acid (VPBA, 98%) were obtained from Sigma (St. Louis, MO, USA). Polystyrene (MW = 2,000,000) was from Alfa Aesar (Shanghai, China). Azobisisobutyronitrile (AIBN, AR) was purchased from Kemiou Chemical Reagent Co., Ltd. (Tianjin, China). Other reagents were of anlytical grade and the crude extracts were purified from Gossypium herbaceam obtained from the Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences. 2.2 Instruments A HPLC system K3800 consisting of a UV2000/2000D UV/Vis detector, a P2000 high-pressure pump, and a K3800 chromatography workstation (Kai'ao Technology Development Co. Ltd, Beijing, China) was used. The detection was performed at 280 nm with a flow rate of 1.0 ml/min. All of the mobile phases were filtered through a 0.22 µm membrane from Millipore before use. Column void volumes were measured by injection of 20 µl of acetone (0.1%, v/v) in the corresponding mobile phase. 2.3 Preparation of poly(VPBA-co-EDMA) monoliths In order to achieve specific meterial with good permeability and selectivity, the monoliths were polymerized under the appropriate concentration of polymer solution,

7

as shown in Table 1. The detailed process of preparing poly(VPBA-co-EDMA) monoliths was as follow: 4-VPBA was used as functional monomer, EDMA as crosslinker monomer, a mixture of polystyrene (PS) solution in tetrahydrofuran (THF) and toluene as porogen. Then the prepolymerization solution was ultrasoniced and introduced into stainless-steel column tube (100 mm×4.6 mm i.d) in 55℃ water bath for 24 h. After polymerization, the column was rinsed by THF and methanol/buffer (pH 4.0) (50:50, v/v) to remove PS and the unreacted reagents. The properties of the monoliths were measured at 77 K by nitrogen adsorption–desorption isotherms using a V-Sorb 2800TP Surface Area and Pore Distribution Analyzer instrument (Gold APP Instruments Corporation China, Beijing, China). 2.4 Separation of hyperoside and isoquercitrin from natural extract The resulting selective material was pumped out from the stainless steel column and ground and sieved with a 41 µm-sieve. Then the uniform granules was packed into a home-made glass column (200 mm×9 mm) with some cotton put in the end of the column for solid phase extraction (SPE). HYP (1 mg) and ISO (1 mg) were dissolved in 0.5 ml methanol and the solution was loaded onto the polyVPBA-EDMA SPE column. Following loading of the mixture sample in methanol and an initial washing with 20 ml of methanol, methanol/water (3:7, v/v) (20 ml) was employed as washing solvent and acetonitrile/buffer (pH 4.0) (3:7, v/v) (20 ml) as eluting solvent. Similarly, 0.5 ml of methanol solution of real sample which was treated by ISO-imprinted polymer [24] was also loaded onto the polyVPBA-EDMA SPE column. The elution solvent was

8

evaporated and the residue was dissolved in 1 mL of water-acetonitrile (50:50) mixture and analyzed by a HPLC system (Thermo Sher, USA) consisting of a quaternary gradient LPG-3400SD pump, a VWD-3100 detector (including a flow cell), and a WPS-3000SL auto sampler. Separation was performed on a Sun Fire TM C18 (250 mm × 4.6 mm, 5 µm) (Waters). 3 Results and discussion 3.1 Preparation of poly(VPBA-co-EDMA) monolithic column 3.1.1 Choice of linear macromolecule The characteristic of macromolecular porogen in preparation of polymer monolith is that it can induce the viscosity enhancing of system, thus the penetration of monomers into the growing nuclei is limited by decreasing diffusion mobility, which increases with increase in its concentration [20]. This viscoelastic phenomenon may decay with a long relaxation time especially in a highly viscous solution. Since polyethylene glycol (PEG) is the most commonly employed linear macromolecule for polymer preparation [6], it was selected as the starting point for the optimization of monolith preparetion for the poly(VPBA-co-EDMA). We also used linear macromolecule PMMA to prepare polyVPBA (data not shown). In present work, it was found that PS other than PMMA or PEG was the optimal macromolecular porogen, maybe due to potential complexation through hydrogen bonding of the free carbonyl groups from PMMA and VPBA. The combination of PS solution in THF and toluene was adopted as porogenic solvent to control the pore morphology of the poly(VPBA-co-EDMA). Depending on relative polarity to the polymer, good solvent

9

(toluene) serves as microporogens to provide high surface areas, while poor solvent (PS solution) acts like macroporogens to permit good column permeability [25]. Using such solvents mixture, a linear dependence of flow rate on column back pressure was observed, indicating good mechanical stability of these monoliths. As expected, the poly(VPBA-co-EDMA) monolith displayed affinity towards HYP (Fig. 2a), cis-diol-containing compound, since reversible covalent reaction between cis-diols and boronic acid ligands can be formed. This can be demonstrated by evaluating on the difference of the chromatographic retention of HYP and ISO. At a basic mobile phase (pH 7.0), ISO had less retention while HYP was strongly retained on the poly(VPBA-co-EDMA) monolith. When the mobile phase was acidic (pH 3.6), little difference in retention of two flavonol glycosides was observed (data not shown). Such a typical behaviour of affinity towards cis-diol molecules as the materials of boronic acid indicated that the extent of reversible formation of cyclic esters depended on the position and number of hydroxyl groups of flavonol glycosides. For comparison, poly(VPBA-co-EDMA) monolith was prepared in porogenc solvent without PS. In the recipe of preparing the monolith, macromolecular porogen (PS solution in THF) was replaced by a mixture of THF and tolune, in which the total volume of the porogenic mixture remained constant. In contrast to those poly(VPBA-co-EDMA)

monoliths

made

in

macromolecule

porogen,

the

poly(VPBA-co-EDMA) monolith (N8) made in the porogen without PS showed too high column pressure when evaluated by HPLC. Thus, the monolith had to be crushed

10

and packed in a flash column (1 ml) and evaluated at very low flow rate by HPLC. As shown in Fig. 2b, a long elution time was needed. Furthermore, selectivity factors of HYP and ISO on the monolith (N8) prepared with the small molecule porogen were lower than the macromolecular porogen, which may be attributed to the difference in morphology of the two monoliths. Multipoint BET measurement was carried out to get meso-pore information of the poly(VPBA-co-EDMA) monolith. Isotherms of “type IV” were observed, which are usually

related

to

mesoporous

materials.

The

hysteresis

loops

of

the

poly(VPBA-co-EDMA) monolith made with PS resembled H1 types [26], while the poly(VPBA-co-EDMA) monolith made without PS indicated the loops of H3 types (Fig. S1), suggesting that porous materials changes from cylindrical pores to slit-shapes pores. In addition, the poly(VPBA-co-EDMA) monolith made in the porogen with PS indicated greater mesopores (pore size of 12.0 nm) than the monoliths prepared in the porogen without PS (pore size of 6.9 nm) (Table 2). 3.1.2 Optimization of PS concentration The influence of PS concentration on the performance of the resultant poly(VPBA-co-EDMA) monolith was studied. Fig. 3a shows a trend of increase in selectivity as PS concentration increased from 10 to 40 mg/ml. This may be attributed to the changes in the thermodynamic quality of the porogenic mixture thanks to the shift in PS concentration, directly affecting the point at which the phase separation occurs during the polymerization process [25]. Further increase of PS concentration was impossible as a result of the limit of PS solubility in THF. The surface area of the

11

resultant poly(VPBA-co-EDMA) monolith was decreased with the increase in PS concentration. Thus, the increase in selectivity was not due to the the shift in surface area of the poly(VPBA-co-EDMA) monolith; otherwise, smaller surface area would cause the decrease in selectivity in virtue of the decline of interaction sites. Fig. S2 indicated the porous structure of poly(VPBA-co-EDMA) monolith obtained using PS of different concentration as solvents under otherwise identical polymerization conditions. The monolith made in the porogen without PS showed the smallest microglobules and no macropore was found. A trend of increase in size of microglobules and macropore was observed as PS concentration increased. The shift in large through pores of the poly(VPBA-co-EDMA) monoliths can be quantitatively characterized by the pore diameter at the maximum of the pore distribution curve (mode pore size) determined with mercury intrusion porosimetry (Fig. S3). As PS concentration increased from 20 to 40 mg/mL, the mode pore size of the monolith increased from 2,200 (N5) to 8,100 nm (N4). Since all the poly(VPBA-co-EDMA) monoliths have the same chemical composition, this morphology difference can be attributed to the result of the diverse solubility parameters of the PS solution with different concentration. 3.1.3 Optimization of ratio of functional monomer to crosslinker The effects of functional monomer to crosslinker (F/C) ratio on the retention factors and selective factors of the resulting poly(VPBA-co-EDMA) monolith were also investigated (Fig. 3b). In the present study, adjusting the amount of crosslinker added to an otherwise constant pre-polymerization mixture varied the molar ratio of

12

VPBA to EDMA. In general, higher VPBA content gave rise to the monolith with higher selectivity because they could provide relatively more phenylboronate functional

groups.

Furthermore,

the

surface

area

of

the

resultant

poly(VPBA-co-EDMA) monolith was decreased sharply with the increase in the ratio of VPBA to crosslinker (Table S1). It was found that the optimum ratio of VPBA to EDMA was 1:2 in terms of the ratio of retention factor for HYP to ISO (17.2) with a compromise of surface area and amount of phenylboronate groups (Fig. 3b), which ratio was close to the work of Chen et al [11]. The pressures applied to the poly(VPBA-co-EDMA) monolith at different F/C ratio were also investigated. The difference in column permeability (9.69×10−8 to 5.86×10−7 mm2) at diferent F/C ratio is shown by the profiles of pressure drop vs flowrate (Fig. S4), which is due to the different size of macropores. The pressure drop of the poly(VPBA-co-EDMA) monolith at lower F/C ratio (1/4) was higher than the poly(VPBA-co-EDMA) monolith at higher F/C ratio (7/3). A deviation from a liner relationship between pressure drop and flow-rate was not observed even at flow rate up to 3 ml/min, demonstrating high rigidity of the poly(VPBA-co-EDMA) monoliths. 3.1.4 Effect of molecular weight of PS To examine effect of varying the molecular weight of PS, a number of poly(VPBA-co-EDMA) monoliths were made using PS with different molecular weight as macromolecular porogen at fixed concentration of PS (40 mg/ml) (Fig. 3c). The porogen in absence of any PS led to the poly(VPBA-co-EDMA) monolith with significanltly high surface area but low column peambility, i.e., high back pressure.

13

Increasing molecular weight of PS produced pores with smaller pore size, while the surface area became higher (Table S2). However, almost no change of surface area (8.0 m2/g) for PS-280,000 replacing PS-150,000 was observed. It should be noted that the relationship between molecular weight of PS and surface area (or pore diameter) was not linear, and therefore, changing the molecular weight from 350,000 to 2,000,000 led to monoliths with almost identical BET surface area (Table S2). Further increase

in

molecular

weights

of

PS

caused

marginal

effects

on

the

poly(VPBA-co-EDMA) monolith, arising from insolubility and the difficulties of removing the PS from the polymer. The shift in retention factors and selective factors of the resulting poly(VPBA-co-EDMA) monolith was also minor. As shown in Fig. 4, as PS concentration increased, the poly(VPBA-co-EDMA) monolith turned morphology from the agglomerated globules to 3D continuous skeletal structures (N9, prepared in PS solution with ultra high molecular weight of 2,000,000). The large PS chain clusters of low miscibility to VPBA and EDMA are presumably separated earlier as macropore forming and the solvent phase (THF) of better miscibility to VPBA and EDMA is separated later as mesopore forming. The latter was proved by preparing poly(VPBA-co-EDMA) monolith in porogenic solvent in absence of any PS. Therefore, the both component in the macromolecular porogenic solution (PS and THF) separated independently without any mutual interaction, i.e., mesopore and macropore formations might proceed independently.

14

3.2 Retention properties and selectivity of poly(VPBA-co-EDMA) monolith Although the poly(VPBA-co-EDMA) monolith allowed for the capture for HYP, the nonspecific adsorption for ISO needed to be considered due to the existence of mechanism of reversed phase retention resulting from the hydrophobic EDMA used [12]. Fig. 5 shows the effect of content of organic modifier on the retention of HYP and ISO. The retention factor of ISO gradually decreased with the reduction of ACN content from 95% to 50%, indicating the existence of a reversed phase mechanism. However, the retention time of HYP changed significantly as the change of ACN content. The results indicated that a mechanism of strong affinity played a leading role in the recognition of HYP. Moreover, the peak response for HYP on chromatograms increased with the increase of ACN content. Herein, 90% ACN was selected as the optimized organic modifier content for further study. To confirm that the retention of HYP on the poly(VPBA-co-EDMA) monolith was indeed via boronate affinity, the pH in mobile phase was varied from 3.6 to 7.0 (Fig. 6). Since VPBA is weak acid (pKa ∼8.6), poly(VPBA-co-EDMA) can offer electrostatic interaction with charged analytes. As expected, the retention factor of HYP and ISO increased with the increase of pH in mobile phase due to its strong ion-exchange interaction with stationary phase [1]. It was observed that the retention factors of HYP sharply increased as the pH value of the buffer increased, with a 300% increase in retention factor from pH 5.6 to 7.0. Considering that at pH 7.0 there was 1.5 pH unit lower than the pKa value of poly(VPBA-co-EDMA), only 3% of the total acid was negatively charged. Therefore, on surface of the polyVPBA tetrahedral

15

phenylboronate anion can be formed [1] due to a weak acid or neutral surrounding of the mobile phase at the pH value. This was consistent with previous results where it has been shown that cyclic anionic ester complexes formed by the phenylboronate groups and HYP are stabilised at higher pH [12]. 3.3 Thermodynamic study The impact of temperature on the retention of HYP and ISO was investigated by shifting temperature from 25◦C to 45◦C on poly(VPBA-co-EDMA) monolith (N1). With column temperature increasing, it was observed that the retention factors of HYP and ISO were increased. Data obtained from the thermodynamic properties of the separation of the two analogues was evaluated by van’t Hoff equation [27]: lnk '  

H RT

ln  



S

H RT

R



 ln

(1)

S (2)

R

ΔH, ΔS, ΔΔH, and ΔΔS can be obtained from the slopes and intercepts of linear portion of the equations (1) and (2). Over the temperature range in our experiment, the van’t Hoff plots were linear for HYP and ISO on the boronate-affinity monolith (N1)(Fig. S5). As shown in Table S3, the value of the apparent ΔH for HYP-poly(VPBA-co-EDMA) interaction (5.5 kJ mol-1) was lower than that for the ISO-poly(VPBA-co-EDMA) interaction (6.8 kJ mol-1). Furthermore, the value of TΔS = 11.3 kJ mol-1 (2.7 kcal mol-1) for HYP was 2 times higher than the value of TΔS (1-1.4 kcal mol-1) calculated accroding to weak complexes [28]. In contrast, the value of TΔS was 5.6 kJ mol-1 (1.3 kcal mol-1) for

16

ISO. This suggested that HYP had stronger interaction to the phenylboronate groups due to boronate affinity than ISO in spite of similar chemical structure. Moreover, |ΔΔH| < T|ΔΔS| indicated that the separation between two analogues on this boronate-affinity monolith was an entropy-controlled process. 3.4 Selectivity of boronate-affinity monolith The selectivity of the poly(VPBA-co-EDMA) monolith (N1) was further evaluated by comparing retention factor of HYP and its structure analogues in acetonitrile/water/acetate (90/9/1, v/v/v) (Fig. 7). Selectivity factor, defined by the ratio of retention factor of HYP to its analogues (isoquercitrin, astragalin, rutin, quercimeritrin and liquiritin) was 16.2, 11.1, 12.1, 11.7 and 11.4, respectively. According to the molecular recognition and selective separation of boronate affinity, the flavonol glycosides without cis-diol structure were not be captured on the boronate-affinity monolith since they are not capable of forming reversible cyclic, anionic esters with the boronate groups. Meanwhile, the secondary interactions can also greatly affect the retention and selectivity of boronate affinity materials, hydrophobic interaction, ionic interaction, hydrogen binding [1]. To demonstrate the selectivity further, the boronate-affinity monolith developed here was applied to SPE. Because of very similar structure, no separation of HYP and ISO can be achieved [24] by highly selective material, molecularly imprinted polymer with ISO imprints, due to cross-reactivity [29]. While N15 proved to be the better polymer in terms of selectivity, N2, the boronate-affinity materials with the smallest selectivity during HPLC evaluation was enough as SPE material to separate HYP and

17

ISO well. After optimisation of the SPE protocols (Fig. S6), it was found that when pure methanol was used as rinsing solvent to wash ISO and its analogues, the most of ingredients were still retained in the SPE column, which can be rinsed by methanol/water (3:7, v/v) further. In non-competitive SPE experiments, the high purify HYP (99.96%) and ISO (100%) were obtained with recovery of 83.7% and 78.6%, respectively. The most of weak retention substance can be rinsed by 10 ml of 20% and 25% methanol aqueous solution, respectively. Then, to demonstrate the potential of the selective extraction for real sample-a diluted extract of cotton flower which mainly contains flavonoids, was also extracted using the boronate-affinity materials. As show in Fig. 8, the boronate-affinity materials for SPE can remove the most of ISO and the interfering compounds, leading to the high recovery of HYP (>80%). TRI, one analogues of HYP having the same structure of cis-diol with very low content in crude extract, can also be obtained with high recovery (85%) after SPE. These results clearly illustrated that the developed polyVPBA is an effective sorbent for the separation and extraction of flavonol glycosides. 4. Conclusion In summary, the possibility to prepare boronate-affinity monolithic column in wide bore tubes was demonstrated by using the porogen of high concentration of PS. The amount of crosslinker was observed to be one of the important factors to affect the selectivity of boronate-affinity monolithic column. We also found that the selectivity of the boronate-affinity monolith can be tailored by varying the molecular weight of PS and the concentration of PS dissolved in the porogen mixture. The meso-pore characterization of the boronate-affinity monolithic column was also found 18

to be dependent on the molecular weight of PS used in the pre-polymerization mixture. In contrast to previous reports, the results displayed that it is possible to tune the pore dimensions of boronate-affinity monolithic column in alternative way. These findings extend the potential applicability of separation and preconcentration from natural products, particularly to the cases where selective separation for cis-diol-containing compounds is required. Acknowledgments This work was supported by supported by the National Natural Science Foundation of China (grant No. U1303202) and the Hundreds Talents Program of the Chinese Academy of Sciences.

19

References [1] D. Li, Y. Chen, Z. Liu, Boronate affinity materials for separation and molecular recognition: structure, properties and applications, Chem. Soc. Rev. 44 (2015) 8097-8123. [2] C. Lü, H. Li, H. Wang, Z. Liu, Probing the interactions between boronic acids and cis-diol-containing biomolecules by affinity capillary electrophoresis, Anal. Chem. 85 (2013) 2361-2369. [3] Y. Liu, Y. Lu, Z. Liu, Restricted access boronate affinity porous monolith as a protein A mimetic for the specific capture of immunoglobulin, Chem. Sci. 3 (2012) 1467-1471. [4] Z. Xu, K.M. Uddin, T. Kamra, J. Schnadt, L.Ye, Fluorescent boronic acid polymer grafted on silica particles for affinity separation of saccharides, ACS Appl. Mater. Interfaces 6 (2014) 1406-1414. [5]

R.

Bandari,

M.R.

Buchmeiser,

Functional

monolithic

materials

for

boronate-affinity chromatography via Schrock catalyst-triggered ring-opening metathesis polymerization. Macromol. Rapid Commun. 33 (2012) 1399-1403. [6] F. Svec, Y. Lv, Advances and recent trends in the field of monolithic columns for chromatography, Anal. Chem. 87 (2015) 250-273. [7] O. G. Potter, M. C. Breadmore, E. F. Hilder, Boronate functionalised polymer monoliths for microscale affinity chromatography, Analyst 131 (2006) 1094-1096. [8] Q. Yang, D. Huang, S. Jin, H. Zhou, P. Zhou, One-step synthesis of an organic–inorganic hybrid boronate affinity monolithic column with synergistic

20

co-monomers, Analyst 138 (2013) 4752-4755. [9] X. Yang, Y. Hu, G. Li, Online micro-solid-phase extraction based on boronate affinity monolithic column coupled with high-performance liquid chromatography for the determination of monoamine neurotransmitters in human urine, J. Chromatogr. A 1342 (2014) 37-43. [10] Z. A. Lin, J. L. Pang,. Y. Lin, H. Huang, Z. W. Cai, L. Zhang, G. N. Chen, Preparation and evaluation of a phenylboronate affinity monolith for selective capture of glycoproteins by capillary liquid chromatography, Analyst 136 (2011) 3281-3288. [11] M. Chen, Y. Lu, Q. Ma, L. Guo, Y.-Q. Feng, Boronate affinity monolith for highly selective enrichment of glycopeptides and glycoproteins, Analyst 134 (2009) 2158-2164. [12] L. Ren, Z. Liu, M. Dong, M. Ye, H. Zou, Synthesis and characterization of a new boronate affinity monolithic capillary for specific capture of cis-diol-containing compounds, J. Chromatogr. A 1216 (2009) 4768-4774. [13] H. Hu, Y. Zhang, Y. Zhang, X. Huang, D. Yuan, Preparation of a new sorbent based on boronate affinity monolith and evaluation of its extraction performance for nitrogen-containing pollutants, J. Chromatogr. A 1342 (2014) 8-15. [14] L. Ren, Z. Liu, Y. Liu, P. Dou, H.-Y. Chen, Ring-opening polymerization with synergistic co-monomers: access to a boronate-functionalized polymeric monolith for the specific capture of cis-diol-containing biomolecules under neutral conditions, Angew. Chem. Inter. Ed. 121 (2009) 6832-6835. [15] C. Cakal, J. P. Ferrance, J P. Landers, P. Caglar, Microchip extraction of

21

catecholamines using a boronic acid functional affinity monolith, Anal. Chim. Acta 690 (2011) 94-100. [16] F. Yang, J. Mao, X. W. He, L. X. Chen, Y. K. Zhang, Preparation of a boronate-functionalized affinity hybrid monolith for specific capture of glycoproteins, Anal. Bioanal. Chem. 405 (2013) 5321-5331. [17] H. Tanaka, Formation of network and cellular structures by viscoelastic phase separation, Adv. Mater. 21 (2009) 1872-1880. [18] X. Wang, M. Okada, C C. Han, Viscoelastic phase separation induced by polymerization

of

n-butyl

methacrylate

in

the

presence

of

poly(dimethylsiloxane-co-diphenylsiloxane), Macromolecules 39 (2006) 5127-5132. [19] H. Aoki, T. Kubo, T. Ikegami, N. Tanaka, K. Hosoya, D. Tokuda, N. Ishizuka, Preparation of glycerol dimethacrylate-based polymer monolith with unusual porous properties achieved via viscoelastic phase separation induced by monodisperse ultra high molecular weight poly(styrene) as a porogen, J. Chromatogr. A 1119 (2006) 66-79. [20]

H.

Aoki,

N.

Tanaka,

T.

Kubo,

K.

Hosoya,

Poly(glycerin

1,3-dimethacrylate)-based monolith with a bicontinuous structure tailored as HPLC column by photoinitiated in situ radical polymerization via viscoelastic phase separation, J. Polym. Sci. A: Polym. Chem. 46 (2008) 4651-4673. [21] J. Seidl, J. Malinsky, K. Dusek, W. Heitz, Macroporous styrene divinylbenzene copolymers and their application in chromatography and for the preparation of ion exchangers, Adv. Polym. Sci. 5 (1967) 113-213.

22

[22] D. Li, Q. Li, S. Wang, J. Ye, H. Nie, Z. Liu, Pyridinylboronic acid-functionalized organic–silica hybrid monolithic capillary for the selective enrichment and separation of cis-diol-containing biomolecules at acidic pH, J. Chromatogr. A 1339 (2014) 103-109. [23] Q. Li, C. Lü, H. Li, Y. Liu, H. Wang, X. Wang, Z. Liu, Preparation of organic-silica hybrid boronate affinity monolithic column for the specific capture and separation of cis-diol containing compounds, J. Chromatogr. A 1256 (2012) 114- 120. [24] X.-J. Li, X.-X. Chen, G.-Y. Sun, Y. X. Zhao, Z.-S. Liu, H. A. Aisa, Green synthesis and evaluation of isoquericin imprinted polymers for class-selective separation and purification of flavonol glycosides, Anal. Methods 7 (2015) 4717-4724. [25] P.-P. Shang, Q.-Q. Pang, L.-S. Zhang, Y.-P. Huang, Z.-S. Liu, Supermacroporous polymer monolith prepared with polymeric porogens via viscoelastic phase separation for capillary electrochromatography, J. Chromatogr. A 1369 (2014) 170-180. [26] K. S. W. Sing, Reporting physisorption data for gas/solid systems with special reference to the determination of surface area and porosity, Pure Appl. Chem. 54 (1982) 2201-2218. [27] S.A. Piletsky, I. Mijangos, A. Guerreiro, E.V. Piletska, I. Chianella, K. Karim, A.P.F. Turner, Polymer cookery: Influence of polymerization time and different initiation

conditions

on

performance

of

molecularly

imprinted

polymers,

Macromolecules 38 (2005) 1410-1414. [28] S. E. Holroyd, P. Groves, M. S. Searle, U. Gerhard, D. H. Williams, Rational

23

design and binding of modified cell-wall peptides to vancomycin-group antibiotics: Factorising free energy contributions to binding, Tetrahedron 49 (1993) 9171-9182. [29] G. Wulff, Enzyme-like catalysis by molecularly imprinted polymers, Chem. Rev. 102 (2002) 1-28.

24

Figure Captions Fig. 1 Structures of flavonol glycosides tested

25

Fig. 2 Comparison of separation of HYP and ISO on poly(VPBA-co-EDMA) monolith made with macromolecule porogen (N1) (a) and small molecule porogen (N8) (b). Mobile phase: acetonitrile/water/acetic acid (90/9/1, v/v/v); detection wavelength: 280 nm; injection: 20 µl; temperature: 30℃. N8 was packed into a balnk flash-column (volume = 1 ml) at flow rate of 0.01 ml/min with backpressure of 0.40 MPa; N1 was monolith at flow rate of 1.0 ml/min with backpressure of 0.10 MPa.

26

Fig. 3 Retention factors and selectivity factors of poly(VPBA-co-EDMA) monolith made with the different concentration of PS (a), ratio of VPBA to EDMA (b), and molecular weight of PS (c). Mobile phase: acetonitrile/water/acetic acid (90/9/1, v/v/v); detection wavelength: 280 nm; flow rate: 1.0 ml/min; injection: 20 µl; temperature: 30℃.

27

Fig. 4 Morphologies characterization of poly(VPBA-co-EDMA) monoliths made with the different molecular weight of PS by surface scanning electron microscope (SEM).

28

Fig. 5 The chromatographic retention behavior of poly(VPBA-co-EDMA) monolith (N1) with various content of acetonitrile. Mobile phase: acetonitrile/water containing 1% acetic acid; detection wavelength: 280 nm; flow rate: 1.0 ml/min; injection: 20 µl; temperature: 30℃.

29

Fig. 6 The retention factor and selectivity on the poly(VPBA-co-EDMA) monolithic column (N1) with various pH. Mobile phase: acetonitrile/buffer (50/50, v/v), 10 mM acetate buffer (pH: 3.6-5.6) and 10 mM phosphate buffer (pH: 5.9 and 7.0); detection wave length: 280 nm; flow rate: 1.0 ml/min; injection: 20 µl; temperature: 30℃.

30

Fig. 7 The selectivity of poly(VPBA-co-EDMA) monolith (N1). Mobile phase: acetonitrile/water/acetic acid (90/9/1, v/v/v); detection wave length: 280 nm; flow rate: 1.0 ml/min; injection: 20 µl; temperature: 30℃.

31

Fig. 8 The crude extracts was purified and separated by poly(VPBA-co-EDMA)(N1) packed in SPE-column. The mobile phase consisted of solvent A (methanol) and solvent B (0.3% phosphoric acid aqueous solution) and solvent D (acetonitrile) with the following gradient: 12% A, 77% B, 11% D, 0-28 min; 12% A, 77-66% B, 11-22% D, 28-60 min; 12% A, 66-65% B, 22 / 23% D, 60 / 90 min. Flow rate, 1.0 ml/min; detection wavelength, 360 nm; injection volume, 10 µl; temperature 30℃.

32

Tables Table 1 Preparation protocol for poly(VPBA-co-EDMA) monoliths. Column no.

VPBA (mg)

Concentration of PS (mg/ml)

Volume of PS (ml)

MW of polymer

Type of polymer

N1

258.95

40

2.25

350,000

PS

142

228.5

40

N2

147.97

40

2.25

350,000

PS

565

228.5

40

N3

73.99

40

2.25

350,000

PS

565

228.5

40

N4

73.99

40

2.25

350,000

PS

377

228.5

40

N5

73.99

20

2.25

350,000

PS

377

228.5

40

N6

73.99

30

2.25

350,000

PS

377

228.5

40

N7

73.99

10

2.25

350,000

PS

377

228.5

40

N8

73.99

0

2.25a

350,000

PS

377

228.5

40

N9

73.99

40

2.25

2,000,000

PS

377

228.5

40

N10

73.99

40

2.25

280,000

PS

377

228.5

40

N11

73.99

40

2.25

150,000

PS

377

228.5

40

N12

73.99

40

2.25

--

PMMA

377

228.5

40

N13

73.99

40

2.25

--

PEG

377

228.5

40

N14

332.91

40

2.25

350,000

PS

142

228.5

40

N15

184.95

40

2.25

350,000

PS

142

228.5

40

N16

110.97

40

2.25

350,000

PS

142

228.5

40

N17

36.99

40

2.25

350,000

PS

142

228.5

40

a

pure THF

33

EDMA Toluene (µl) (µl)

AIBN (mg)

Table 2 The pore parameters of poly(VPBA-co-EDMA) monoliths made with different content of PS Number

Content of PS (mg/ml)

Surface area (m2/g)

Pore volume (cm3/g)

Pore size (nm)

N4

40

58.9

0.075

5.9

N6

30

154.3

0.13

3.5

N5

20

223.9

0.31

6.8

N7

10

335.8

0.70

12.0

N8

0

324.3

0.46

6.9

34