13646
Langmuir 2008, 24, 13646-13652
An Effective Way To Hydrophilize Gigaporous Polystyrene Microspheres as Rapid Chromatographic Separation Media for Proteins Jian-Bo Qu,†,‡ Wei-Qing Zhou,† Wei Wei,†,‡ Zhi-Guo Su,† and Guang-Hui Ma*,† National Key Laboratory of Biochemical Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China, and Graduate UniVersity of Chinese Academy of Sciences, Beijing 100049, China ReceiVed May 14, 2008. ReVised Manuscript ReceiVed July 14, 2008 To overcome the disadvantages of protein denaturation and nonspecific adsorption on poly(styrene-divinylbenzene) (PS) medium as a chromatographic support, gigaporous PS microspheres prepared in our previous study were coated with hydrophobically modified agarose (phenoxyl agarose, Agap). Both the modification of agarose and the gigaporous structure of PS microspheres provided an advantage that facilitated the coating of Agap onto PS microspheres. The amount of Agap adsorbed onto the PS surface was examined as a function of the polymer concentration, and various samples of microspheres, differing in surface Agap density, were prepared. The adsorbed layer was then stabilized by chemical cross-linking and its stability was evaluated in the presence of sodium dodecyl sulfate. Results showed that PS microspheres were successfully coated with Agap, while the gigaporous structure could be well maintained. After coating, the nonspecific adsorption of proteins on PS microspheres was greatly reduced. Flow hydrodynamics experiments showed that the Agap-co-PS column had low backpressure, good permeability, and mechanical stability. Such a procedure could provide a hydrophilic low-pressure liquid chromatographic support for different types of chromatography, since the Agap layer may be easily derivatized by classical methods, and because of their good permeability, the coated microspheres have great potential applications in high-speed protein chromatography.
1. Introduction The interest in and demand for biomacromolecules, such as proteins and plasmid DNA in biotechnology, biochemistry, and medicine industries, have contributed to an increased exploitation of effective bioseparation medium in chromatography. The most popular medium to date, agarose, is limited by its lack of mechanical rigidity, even in highly cross-linked varieties, thus restricting its application in large-scale operations. It is imperative that large-scale liquid chromatographic media should be mechanically strong, as lightly cross-linked semirigid polymer networks are easily compressed and can only be operated under low to medium pressure. Ideally, a “base support” should be (1) chemically and mechanically stable, (2) hydrophilic, (3) charge free, (4) easily derivatizable,1 and (5) free of a stagnant mobile phase mass transfer problem. Poly(styrene-divinylbenzene) (PS) microspheres are of increasing interest over silica and conventional separation media (e.g., dextran and agarose) as chromatography packing materials for proteins and antibodies owing to their excellent mechanical properties and a good chemical stability over a wide pH range.2 However, the pore size of most conventional porous PS particles is in the range of 10-30 nm, which has the problem of a longer time when separating these bioproducts due to the slow diffusion rate through the interior of the stationary phase particles.3,4 Ideal polymeric particles with large pore size (larger than 10 or 20 times the solute molecule size) can overcome the shortcoming * Corresponding author. Fax: +86 10-82627072. E-mail: ghma@ home.ipe.ac.cn. † Chinese Academy of Sciences. ‡ Graduate University of Chinese Academy of Sciences.
(1) Yang, Y. B.; Regnier, F. E. J. Chromatogr. 1991, 544, 233–247. (2) Lee, D. P. J. Chromatogr. Sci. 1982, 20, 203–208. (3) Afeyan, N. B.; Regnier, F. E; Dean, R. C. Perseptive Biosystems Inc., U.S. Patent 5,833,861, 1998. (4) Sherrington, D. C. Chem. Commun. 1998, 2275–2286.
mentioned above. Our research group has developed an easy, novel method to prepare gigaporous PS microspheres with pore size of about 300-500 nm,5 which can be used as a perfusion chromatographic support.6-11 Unfortunately, the native PS beads are limited in the chromatography for protein separation due to their highly hydrophobic properties, which will lead to nonspecific adsorption and denaturation of proteins. The use of water-soluble polymers (e.g., amphiphilic polysaccharides,12-14 poly(ethylene oxide)poly(propylene oxide) copolymers,15-18 poly(vinyl alcohol)19-21), which have low concentration of hydrophobic groups acting as anchors, in the adsorbed coating technology turns out (5) Zhou, W. Q.; Gu, T. Y.; Su, Z. G.; Ma, G. H. Polymer 2007, 48, 1981– 1988. (6) Afeyan, N. B.; Fulton, S. P.; Regnier, F. E. J. Chromatogr. 1991, 544, 267–279. (7) Fulton, S. P.; Afeyan, N. B.; Gordon, N. F.; Regnier, F. E. J. Chromatogr. A 1991, 54, 452–456. (8) McCoy, M.; Kalghatgi, K.; Regnier, F. E.; Afeyan, N. B. J. Chromatogr. A 1996, 743, 221–229. (9) Whitney, D.; McCoy, M.; Gordon, N.; Afeyan, N. B. J. Chromatogr. A 1998, 807, 165–184. (10) Liu, A. L.; Zhu, B. F. Chin. Biotechnol. 2002, 12, 24–25. (11) Xiong, B. H.; Wang, J. D.; Chin, J. Chromatography 1997, 15, 486–489. (12) Malmsten, M.; Lindman, B.; Holmberg, K.; Brink, C. Langmuir 1991, 7, 2412–2414. (13) Malmsten, M.; Tiberg, F. Langmuir 1993, 9, 1098–1103. (14) Fournier, C.; Leonard, M.; Le Coq-leonard, I.; Dellacherie, E. Langmuir 1995, 11, 2344–2347. (15) Shar, J. A.; Obey, T. M.; Cosgrove, T. Colloids Surf. A 1998, 136, 21–23. (16) Schroen, C.G. P. H.; Cohen Stuart, M. A.; van der Voort Maarschalk, K.; van der Padt, A.; van’t Riet, K. Langmuir 1995, 11, 3068–3074. (17) Cassidy, O. E.; Rowley, G.; Fletcher, I. W.; Davies, S. F.; Briggs, D. Int. J. Pharm. 1999, 182, 199–211. (18) Lee, J. H.; Kopeckova, P.; Kopecek, J.; Andrade, J. D. Biomaterials 1990, 11, 455–464. (19) Tuncel, A.; Denizli, A.; Purvis, D.; Lowe Chris, R.; Piskin, E. J. Chromatogr. 1993, 634, 161–168. (20) Leonard, M.; Fournier, C.; Dellacherie, E. J. Chromatogr. B 1995, 664, 39–46. (21) Nash, D. C.; McCreath, G. E.; Chase, H. A. J. Chromatogr. A 1997, 664, 53–64.
10.1021/la801486t CCC: $40.75 2008 American Chemical Society Published on Web 11/04/2008
Hydrophilized Gigaporous Polystyrene Microspheres
Langmuir, Vol. 24, No. 23, 2008 13647
Scheme 1. Fixation of Phenoxy Groups on Agarose
to be one of the most efficient methods for the hydrophilization of PS particles. Furthermore, the adsorbed polymer layer can be chemically cross-linked and thus stabilized. Up to now most previous works were concerned with coating of hydrophobic plate materials or nanoparticles, while few studies on the adsorption of hydrophilic polymers onto porous particles are reported in the literature.14,21 The main challenge was that the pore channels of conventional particles were easily blocked and hydrophilic macromolecules could not conveniently enter inner pores. On the contrary, the large pores of gigaporous PS microspheres in the present study could facilitate the coating with hydrophilic macromolecules. Agarose, a hydroxyl-rich and very biocompatible natural polymer, not only can effectively shield the PS surface from hydrophobic interaction but also provide an easily derivative layer. To the best of our knowledge, hydrophilization of hydrophobic porous microspheres by coating agarose has not yet been reported. There are two main reasons: (1) the average molecular weight of agarose is too large (usually more than 10 kDa) to maintain the porous structure of conventional microspheres, and (2) the adsorption forces between hydrophobic materials and agarose chains are very weak, so it is difficult to cover the interior surface completely with agarose. In this paper, we reported an effective way to hydrophilize gigaporous PS microspheres by adsorption of a hydrophobically modified agarose (Agap). The gigaporous structure of PS microspheres can overcome the problem that pore channels of conventional particles were easily blocked. The effects of the adsorption and crosslinking conditions on the amount of adsorbed Agap and on the stability of polymeric layer were investigated. Fourier transform infrared (FT-IR) spectra, X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), mercury porosimetry measurements (MPM), optical contact angle measuring device (OCA), flow hydrodynamics, laser scanning confocal microscope (LSCM), and bovine serum albumin (BSA) adsorption experiments were used to test the feasibility of the modified microspheres as a perfusion chromatographic base support.
2. Expermental Section 2.1. Materials. The gigaporous poly(styrene-divinylbenzene) microspheres used in this study were synthesized by a modified suspension polymerization processes developed in our previous study.5 The specific surface area was 24.65m2/g, the average diameter was 55 µm (30-85 µm range), and the average pore size was 300 nm (100-500 nm range). The microspheres were extracted with acetone in a Soxhlet extractor for 24 h and then dried under vacuum at 50 °C for further process. Agarose powder was purchased from Promega Corp. (purity >99%); Blue Dextran 2000 and bovine serum albumin (BSA) were obtained from Amresco; glycidyl phenyl ether (GPE; AR) was ordered from WUXI resin factory; epichlorhydrin (EPCl; AR) and ethylene glycol diglycidyl ether (EDGE; AR) were from Sinopharm Chemical Reagent Co., Ltd. All reagents were used as received.
2.2. Preparation of Hydrophobically Modified Agarose (Agap). The introduction of aromatic groups on agarose chains was performed according to Scheme 1. Typically, agarose (2 wt %, 100 mL) was allowed to react with GPE in 1 M NaOH at 60 °C for 48 h. The crude products were then precipitated twice with ethanol at room temperature, dialyzed against deionized water for 48 h, and freezedried. 2.3. Determination of Phenoxy Content of Agap. The phenoxy content of the Agap was determined with a UV spectrophotometer (Ultrospec2100 pro, Amershan Biosciences), by the absorbance at 269 nm.14 The absorbance was measured at 75 °C to keep the Agap solution as a transparent liquid and give a precise phenoxy content of Agap. 2.4. Determination of the Molecular Weight of Agarose before and after Modification. Gel size-exclusion chromatography, in conjunction with online refractive index detector (OPTILAB DSP, Wyatt Technology Co.) and multiangle laser light scattering detector (DAWN EOS, Wyatt Technology Co.), was performed on a TSK G4000sw column at 55 °C, using 0.04 mol/L phosphate buffer (pH 7.3) as mobile phase. The flow rate was 0.5 mL/min, the concentration of the sample solution was 2 mg/mL, and the injection amount was 100 µL. 2.5. Viscometric Measurements of Agarose and Agap Solution. The viscosities of agarose and Agap solution were measured by an L-90 rheometer (Tongji University Machine & Electron Factory) at 55 °C. The concentration of the sample was 20 mg/mL. 2.6. Coating of Gigaporous PS Microspheres with Agap (Agap-co-PS). The hydrophilization of gigaporous PS microspheres was achieved by adsorption of Agap in deonized water (70 mL of solution/g of dry PS microspheres at various Agap concentrations) at 55 °C for 24 h. The suspension was filtered through a sintered glass funnel. Then the microspheres in the funnel were washed with hot water (70 °C, 280 mL/g of dry PS microspheres). The filtrate was collected together for determination of the equilibrium concentration of solution and the nonadsorbed Agap by UV spectroscopy at 269 nm. The same steps without adding PS microspheres were performed as a control experiment to determine the amount of Agap bound to the funnel. It was confirmed that there was no apparent change of Agap concentration in the process. The Agap-co-PS microspheres were then chemically cross-linked with EPCl or EDGE at various concentrations in 0.4 M NaOH (35 mL/g dry PS microspheres) at 25 °C. Finally, the particles were washed with 2% sodium dodecyl sulfate solution (SDS) at 50 °C for 24 h. The final amount of Agap adsorbed onto the PS microspheres was determined as the difference between the initial amount of Agap added and the polymer remaining in all the filtrate, supernatants, and washing solutions. In SDS solutions, Agap was determined according to the anthrone method, by UV spectroscopy at 625 nm.22 The average value of at least triplicate samples are presented for all data. In addition, the coatings adsorbed onto PS microspheres were crosslinked in advance before further characterization and analyzation. 2.7. Composition and Morphology Analysis. XPS (VG Scientific ESCALab220i-XL) was employed to study the composition of PS microspheres before and after coating with Agap. The composition changes between original and coated microspheres were also determined using FT-IR spectroscopy (JASCO FT/IR-400/600). (22) Scott, T. A.; Melvin, E. H. Anal. Chem. 1953, 25(11), 1656–1661.
13648 Langmuir, Vol. 24, No. 23, 2008
Qu et al.
Table 1. Phenoxy Content of Various Agap Agap sample
phenoxy content (mmol/g)
degree of substitution (DS)a
#10 #12 #13 #14 #15 #16 #18 #19 #20
0.1917 0.2916 0.2393 0.3619 0.3361 0.4002 0.107 0.5546 0.598
6.04 9.33 7.6 11.7 10.8 13.0 3.33 18.5 20.1
a
DS: the number of phenoxy groups per 100 agarose units.
Figure 1. Effect of DS on Agap molecular weight and solution viscosity.
Figure 2. Adsorption isotherm of Agap on gigaporous PS microspheres.
The gigaporous structures of PS and Agap-co-PS microspheres were observed by scanning electron microscopy (SEM, JEM-6700F, JEOL). Mercury porosimetry measurements (MPM) were conducted by an AutoPore IV 9500 mercury porosimetry (Micromeritics) to study the difference between PS and Agap-co-PS microspheres. 2.8. Mechanical Stability and Permeability of Medium. The microspheres before and after modification were packed into a stainless steel column (250 × 4.6 mm i.d.) by slurry packing method on a Waters 600E-2487 system (600 pump and controller, dual λ absorbance detector). The mechanical stability and permeability of medium were evaluated through the effect of flow rate on the backpressure of column, and the mobile phase was 20% aqueous ethanol solution. The bed voidage (ε) values were obtained by using Blue Dextran 2000 as a tracer, which reflected the void volume of the column. Blue Dextran 2000 was detected at 260 nm. The flow rate was 1.0 mL/min, the concentration of tracer solution was 6 mg/mL, and the injection amount was 20 µL. For a column packed with porous media, the bed hydraulic permeability (K) can be described by the Blake-Kozeny equation.23
K)
dp2ε3 150(1 - ε)2
(1)
where dp is the particle size (µm) and ε the bed voidage. Also, K can be evaluated by the Dacy law in a laminar flow region.24
K)
µuL ∆P
(2)
where µ is the viscosity of the mobile phase (Pa · s), u the superifical velocity (cm/s), L the length of column (cm), and ∆P the column pressure-drop (Pa). Therefore, K is given from eq 1 and the column pressure-drop can be predicted. 2.9. Evaluations of Surface Hydrophilicity. An optical contact angle measuring device (OCA20, Dataphysics, Germany) was employed to study the surface hydrophilicity of gigaporous PS microspheres before and after coating. The dynamic contact angle of microspheres was measured by using the pressed-disk technique. The microspheres were first pressed into a disk by a mold, and then the water was dropped onto the disk by a needle at 0.5 µL/drop. The test frequency was fixed to 25 times per second. The temperature and relative humidity was 20 °C and 30%, respectively. BSA adsorption was also performed to evaluate the surface hydrophilization efficiency of microspheres after modification.25,26 BSA was dissolved in phosphate-buffered saline (PBS) buffer (pH 7.4). In each experiment, 50 mg of microspheres was incubated with 5.0 mL of BSA solution at different concentrations. The BSA solution without added microspheres was used as control experiment. Typical batch experiments were conducted in a 25 °C water bath, with stirring for 24 h at 120 rpm. At the end of adsorption equilibrium, the microspheres were separated from the solution by centrifugation. The residual protein concentrations (equilibrium concentration) were determined by UV spectroscopy at 280 nm. The protein adsorption amount was determined by the mass balance of proteins. The adsorption of BSA on GE Sepharose 6FF beads was also evaluated as a comparison experiment (BSA concentration, 6 mg/mL). FITC-labeled BSA adsorption experiments were also conducted in a 25 °C water bath. After 24 h adsorption, the microspheres were separated from the FITC-BSA solution by centrifugation and washed with PBS buffer twice to remove desorbed protein molecules completely. After being placed on a slide glass and being covered with a cover glass, the microspheres were observed with a TCS SP2 LCSM (Leica) to visualize the adsorption of FITC-BSA onto microspheres. The samples were excited at 488 nm and the fluorescent images at 520-550 nm wavelengths were then taken.
3. Results and Discussion 3.1. Effect of DS on Agap Molecular Weight and Solution Viscosity. The phenoxy content of the various Agap is quoted in Table 1. According to the phenoxy content of Agap, the degree of substitution (DS) can be calculated. Figure 1 shows the effect of DS on Agap molecular weight and Agap solution viscosity. After modification, the molecular weight of agarose first decreased significantly (from 134 900 to around 24 000) and then increased slightly when DS increased further from 6.04. The former indicated that the scission of agarose chains occurred under base reactive conditions (1 M NaOH, 60 °C); the latter was due to the increase of phenoxy groups on Agap chains. Obviously, the reduction of agarose molecular weight provided an advantage that the coating of Agap onto PS microspheres would be facilitated. (23) Bird R. B.; Stewart W. E.; Lightfoot E. N. Transport Phenomena; WileyVCH, New York, 1960. (24) De la vega, R. M.; Chenou, C.; Loureiro, J. M.; Rodrigue, A. E. Biochem. Eng. J. 1998, 1, 11–23. (25) Leonard, M.; Fournier, C.; Dellacherie, E. J. Chromatogr. B 1995, 664, 39–46. (26) Yoon, J. Y.; Kim, J. H.; Kim, W. S. Colloids Surf. A 1999, 153, 413–419.
Hydrophilized Gigaporous Polystyrene Microspheres
Langmuir, Vol. 24, No. 23, 2008 13649
Figure 3. Characteristic C1s XPS curves for (a) PS and (b) #16 Agap-co-PS microspheres (coating amount is 78.96 mg/g dry microspheres). Table 2. Peak Area Ratio of C1s Curves for PS and Agap-co-PS Microspheres C1s
peak area ratio
sample
peak 1
peak 2
peak 3
peak 2/peak 1
peak 3/peak 1
PS Agap-co-PS
C-C/C-H C-C/C-H
C-OH/C-O-C C-OH/C-O-C
O-CdO O-CdO
0.135 0.707
0.022 0.024
Compared with original agarose solution, the viscosity of Agap solution also showed a great reduction in Figure 1. There were probably two reasons. One was that the molecular weight of agarose decreased significantly after modification. The other was the existence of phenoxy groups, which could reduce the hydrogen bond effect between agarose chains and disentangle the agarose chains. Furthermore, the intramolecular hydrophobic interaction between phenoxy groups would lead to a shrinking conformation in a dilute solution. As can be seen from Figure 1, the viscosity of Agap solution only decreased slightly with the increase of DS (from 6.04 to 20.1). This indicated that the main reason for the abrupt decrease of Agap solution viscosity is the decrease of Agap molecular weight. 3.2. Effect of DS on the Amount of Agap Coated onto PS Microspheres. The coating amount of various Agap on gigaporous PS microspheres is shown in Figure 2. The adsorption isotherm of Agap obtained in water reveals typically Langmuir type where the adsorbed amount rapidly increases with polymer concentrations up to a plateau region. Each plateau was obtained for an equilibrium concentration around 3 mg/mL. It also can be observed from Figure 2 that there was an increase in the amount of Agap adsorbed as DS increased. This was mainly because the adsorption forces between Agap and PS surface are hydrophobic forces, and phenoxy groups worked as hydrophobic anchors in the adsorption process. 3.3. CompositionVerificationoftheAgap-co-PSMicrospheres. XPS C1s spectra of gigaporous PS and Agap-co-PS microspheres are shown in Figure 3. Though both the C1s curve fit results had three peaks, they had different attribution and peak area ratio (see Table 2). For C1s curve fit results of PS microspheres (Figure 3a), the additional two peaks (C-O-C/C-OH peak near 287 eV, O-CdO peak near 289 eV) besides the C-C/C-H peak (284.8 eV) could be attributed to the presence of residual sorbitan monooleate (Span80) in microspheres, which had been integrated into the microspheres during polymerization and could not be extracted by solvent. As C1s curve fit results of Agap-co-PS (Figure 3b) were concerned, peak area ratio of peak 2 (286.6 eV)/peak 1 (284.8 eV) increased greatly (from 0.135 to 0.707) compared with PS microspheres, confirming the coating of Agap onto PS microspheres. FT-IR spectra for PS and Agap-co-PS microspheres are shown in the Supporting Information (Figure S1), which also confirmed that Agap had been successfully coated onto PS microspheres.
3.4. Properties of Pore Size and Microstructure before and after the Modification. Figure 4shows SEM images of the native PS, agarose-co-PS and Agap-co-PS microspheres. When agarose was directly adsorbed onto PS microspheres, the coating amount was not satisfied. Notably, the coating was nonhomogeneous and agarose chains could plug the pore channels of PS microspheres. As the adsorption of Agap was concerned, we found the same phenomena still existed when DS was lower (Figure 4c,d). With the increase of DS (Figure 4e,f), the aggregation of Agap on the surface of PS microspheres disappeared, and the coating became homogeneous. These results indicated that higher phenoxy content helped to maintain the gigaporous structure of particles and homogeneousness of the coating, which is desired as a chromatographic support. When DS was further increased, the Agap would not dissolve well in water as the result of the high hydrophobicity of Agap. Figure 5 shows that pores with the diameter between 100 and 600 nm occupied most of the pore volume in the medium and the greatest incremental pore volume occurred at a pore diameter of around 300 nm. From 100 to 600 nm the incremental pore volumes are almost equal in both medium types, indicating that the macroporous structure of PS microspheres was well maintained after modification. 3.5. Effect of Cross-Linking Conditions on the Stability of Coating. Figure 6 shows the stability of coatings after crosslinking and SDS treatment. Compared with uncross-linked sample, cross-linked samples presented a sharp decrease in the total amount of relative desorbed Agap. When using EPCl as a cross-linking agent, the amount of desorbed Agap decreased with the increase of EPCl concentration due to the increase of cross-linking density. On the contrary, the amount of desorbed Agap after SDS treatment did not show the same tendency. Probably, the partial hydrolyzation of the chloromethyl on excess EPCl would add CHCHOHCH2OH groups on Agap chains, which decreased the hydrophobic effects between Agap and the PS surface, resulting in the facile desorption with SDS treatment. As cross-linking agent EDGE was concerned, the amount of desorbed Agap presented the same tendency as that of EPCl when the EDGE concentration was lower than 0.64 M. However, the tendency was the opposite when the EDGE concentration was higher than 0.64 M, owing to the competitive adsorption between Agap and EDGE. It is interesting that there was no dependence on EDGE concentration when the coating was treated
13650 Langmuir, Vol. 24, No. 23, 2008
Qu et al.
Figure 4. SEM images of gigaporous PS microspheres before and after adsorption agarose and Agap (a, native PS microspheres; b, agarose-co-PS microspheres; c, #18 Agap-co-PS microspheres; d, #13 Agap-co-PS microspheres; e, #16 Agap-co-PS microspheres; f, #19Agap-co-PS microspheres).
Figure 5. Pore size distribution curves of gigaporous PS microspheres before and after coating.
with SDS. We supposed that EDGE with C8 chain would not increase the hydrophilicity of Agap even when hydrolyzed at one end. Taken together, the optimum cross-linking condition
was EDGE as cross-linking agent (D3 group, the total amount of relative desorbed Agap was 8.33%). Finally, the stability of the coating was evaluated under both acidic and basic conditions, and no apparent leakage of Agap was detected after 24 h at ambient temperature either in 1 M HCl or 1 M NaOH (detection limit 0.02 mg of Agap/g of PS microspheres). 3.6. Dynamic Contact Angle of Microspheres Surface before and after Coating. Figure 7 shows the dynamic contact angle curves for native and coated gigaporous PS microspheres surface. After coating, the contact angle of all samples decreased significantly. This indicated that the hydrophilicity of microspheres surface increased greatly. As can also be seen from Figure 7, the contact angle of microspheres after coating decreased first and then rose again with the increase of DS, indicating there are optimum numbers of phenoxy groups on Agap chains. When DS was lower (lower than 13.0), the Agap chains could not cover (27) Sun, G. Y.; Shi, Q. H.; Sun, Y. J. Chromatogr. A 2004, 1061, 159–165. (28) Shi, Y.; Dong, X. Y.; Sun, Y. Chromatographia 2002, 55, 405–410.
Hydrophilized Gigaporous Polystyrene Microspheres
Figure 6. Relative Agap desorption from #19-5 Agap-co-PS sample (77.48 mg of Agap/g of PS microspheres) after cross-linking and SDS treatment. N was uncross-linked sample. E1, E2, E3, E4, and E5 were cross-linked with EPCl (the concentration was 0.13, 0.26, 0.64, 1.02, and 1.53 M, respectively). D1, D2, D3, and D4 were cross-linked with EDGE (the concentration was 0.13, 0.32, 0.64, and 0.93 M, respectively).
Figure 7. Dynamic contact angle of PS microspheres surface before and after coating. Lines 1, 2, 3, 4, 5, and 6 represent PS, #18 Agap-co-PS, #13 Agap-co-PS, #19 Agap-co-PS, #14 Agap-co-PS, and #16 Agapco-PS microspheres, respectively (the DS values for #18, #13, #14, #16, and #19 Agap were 3.33%, 7.60%, 11.7%, 13.0%, and 18.5%, respectively). Table 3. Bed Voidage and Hydraulic Permeability of Column matrix
native PS microspheres
#16 Agap-co-PS
ε Ka (m2) Kb (m2)
0.5096 1.11 × 10-11 3.38 × 10-10
0.5002 1.01 × 10-11 2.88 × 10-10
a Calculated by eq 1. b Calculated by eq 2. The column pressure-drop was taken from experimental data.
the PS surface sufficiently and the coating layer were loosely packed. For the higher DS (higher than 18.5), the coating was packed tightly, but excess phenoxy groups would be exposed on the coating layer. Correspondingly, the hydrophilicity of PS surface would decrease again. 3.7. Flow Hydrodynamics. The mechanical stability and permeability of medium were evaluated for PS gigaporous microspheres before and after modification by the relationship between the flow rate and backpressure of the column. The results showed that the backpressure of PS and Agap-co-PS column under 3612 cm/h was only 1.65 and 2.07 MPa, respectively. The low backpressure of the columns was evidence for the presence of flow-through pores, which reduced the flow resistance. The same phenomenon has been reported with other macroporous
Langmuir, Vol. 24, No. 23, 2008 13651
Figure 8. Adsorption isotherms of BSA on gigaporous PS microspheres before and after coating.
beads.27-29 Detailed experimental data could be found in the Supporting Information(Figure S2). Table 3 gives the bed voidage and hydraulic permeability of column. It is evidently that K values determined by the Blake-Kozeny equation were far smaller than the test data for both columns. Correspondingly, the column pressure-drop data predicted by the Blake-Kozeny equation were far larger than experimental data (see Figure S2 of the Supporting Information). This indicated that the mobile phase in the column flowed not only through the spaces between matrix but also through the gigapores of particles, again confirming the existence of through pores in gigaporous microspheres. 3.8. Adsorption of BSA on the Surface of Microspheres before and after Coating. The adsorption isotherms of BSA on microspheres before and after coating are shown in Figure 8. The BSA amount absorbed on microspheres decreased sharply after coating, which was ascribed to the surfaces of PS microspheres masked well by hydrophilic agarose chains. The adsorbed amount of BSA on #16 Agap-co-PS sample at the plateau (1.18 mg/g of dry microspheres) was 75.8 times lower than that of PS microspheres (89.55 mg/g of dry microspheres). Notably, the adsorbed amount of proteins on Agap-co-PS microspheres surface decreased first and then increased again with the increase of phenoxy content, which has the same tendency as the contact angle of Agap-co-PS microspheres. That is, there was a consistency with the protein adsorption and the surface hydrophobicity of the microspheres. In the control experiment, the adsorption amount of BSA on agarose beads was 0.574 mg/g dry beads, which also indicated that the hydrophilicity of #16 Agap-co-PS microspheres was very close to that of agarose beads. The FITC-BSA adsorption experiment gave a straight insight on the increase of hydrophilicity of PS microspheres surfaces after modification. Detailed results and discussion can be found in the Supporting Information (Figure S3).
4. Conclusions A potential hydrophilic packing support has been developed by coating gigaporous PS microspheres with modified agarose. After modification, the molecular weight of agarose decreased significantly, which provided an advantage that the coating of Agap onto PS microspheres was facilitated. Also, the introduction of phenoxy groups on agarose leads to a derivative that can be adsorbed on the polystyrene porous surface by means of hydrophobic sites. Then the adsorbed layer can be stabilized by chemical cross-linking. After coating, the hydrophilicity and biocompatibility of gigaporous PS microspheres were greatly
13652 Langmuir, Vol. 24, No. 23, 2008
enhanced. The nonspecific adsorption of proteins was nearly reduced to zero through the optimization of both the phenoxy content and the cross-linking conditions of adsorbed layer. Moreover, convective flow of mobile phase through the gigaporous pores in the medium has been confirmed by its low backpressure and the Blake-Kozeny equation. This study provided a hydrophilic chromatographic support for different types of chromatography such as ion-exchange or affinity chromatography, since the agarose coatings may be easily derivatized by classical methods. After further derivation, the coated gigaporous microspheres would be a potential stationery phase in high-speed preparative protein chromatography.
Qu et al.
Acknowledgment. We would like to thank the financial support from the National Natural Science Foundation of China (Nos. 20536050, 20636010, and 20221603) and the ministry of Science and Technology of China (No. 2007CB714305). Supporting Information Available: FT-IR spectra, flow hydrodynamic analysis of packed column, and LCSM images. This material is available free of charge via the Internet at http://pubs.acs.org. LA801486T (29) Shi, Y.; Sun, Y. Chromatographia 2003, 57, 29–35.
