Anal. Chem. 2004, 76, 7013-7022
High-Performance Affinity Monolith Chromatography: Development and Evaluation of Human Serum Albumin Columns Rangan Mallik, Tao Jiang, and David S. Hage*
Chemistry Department, University of Nebraska, Lincoln, Nebraska 68588-0304
Several immobilization methods were explored for the preparation of high-performance affinity monolithic columns containing human serum albumin (HSA). These monoliths were based on a copolymer of glycidyl methacrylate and ethylene dimethacrylate. In one method, the epoxy groups of this copolymer were used directly for the immobilization of HSA through its amine residues (i.e., the epoxy method); in other approaches, these epoxy groups were converted to diols for later use in the carbonyldiimidazole, disuccinimidyl carbonate, and Schiff base methods. Each HSA monolith was evaluated in terms of its total protein content and its retention of several model compounds, including (R/S)-warfarin and D/Ltryptophan. The greatest amount of immobilized HSA was obtained by the Schiff base method, whereas the epoxy method gave the lowest protein content. The Schiff base method also gave the best resolution in chiral separations of (R/S)-warfarin and D/L-tryptophan. All of the immobilization methods gave similar relative activities for HSA in its binding to (R)- and (S)-warfarin, but some differences were noted in the activity of the immobilized HSA for D- and L-tryptophan. The efficiency of these monoliths was found to be greater than that of silica-based HSA columns for (R/S)-warfarin (i.e., analytes with high retention), but little or no difference was seen for D- and L-tryptophan (analytes with weak retention). High performance affinity chromatography (HPAC) is a powerful tool for the separation and analysis of chemicals.1-4 This method uses a HPLC column containing an immobilized ligand capable of specifically binding the analyte or group of analytes of interest. This approach can be used in the study of biological interactions, biomolecule purification, and chiral separations.5-7 * To whom correspondence should be addressed. Phone: 402-472-2744. Fax: 402-472-9402. E-mail:
[email protected]. (1) Hage, D. S.; Austin, J. J. Chromatogr., B 2000, 739, 39. (2) Hage, D. S. J. Chromatogr., B 2002, 768, 3. (3) Larive, C. K.; Lunte, S. M.; Zhong, M.; Perkins, M. D.; Wilson, G. S.; Gokulrangan, G.; Williams, T.; Afroz, F.; Schoeneich, C.; Derrick, T. S.; Middaugh, C. R.; Bogdanowich-Knipp, S. Anal. Chem. 1999, 71, 389R. (4) Lillehoj, E. P.; Malik, V. S. Adv. Biochem. Eng. Biotechnol. 1989, 40, 19. (5) Liu, Y.; Zhao, R.; Shangguan, D.; Zhang, H.; Hongwu, L.; Liu, G. J. Chromatogr., B 2003, 792, 177. (6) Chattopadhyay, A.; Tian, T.; Kortum, L.; Hage, D. S. J. Chromatogr., B 1998, 715, 183. (7) Yang, J.; Hage, D. S. J. Chromatogr. 1993, 645, 241. 10.1021/ac049001q CCC: $27.50 Published on Web 11/04/2004
© 2004 American Chemical Society
Advantages of HPAC include its high specificity, speed, ease of automation, and ability to use the same ligand for multiple applications. An important factor in the success of HPAC or any type of affinity chromatography is the way in which the ligand is attached to the chromatographic support.8-10 Ideally, this immobilized ligand should mimic the same ligand in its native state. This makes the selection of immobilization conditions a key item to consider in the development of any affinity column. Most current HPAC methods use diol-bonded silica or a similar glycol-containing material for the immobilization of ligands.11 However, these materials can have slow mass-transfer properties and significant column back pressures.12,13 This has led to growing interest in alternative supports, such as monolithic columns.14 Monolithic columns have recently been used in reversed-phase methods for the separation of peptides and proteins;15-18 however, there has been little work reported in the use of monolithic columns for affinity chromatography.19,20 One material that can be used for this purpose is a copolymer of glycidyl methacrylate and ethylene dimethacrylate (GMA/EDMA).21 Immobilization of proteins on modified matrixes containing GMA has been described previously.22-27 This can be converted into a diol-bonded (8) Turkova, J. Affinity Chromatography; Elsevier: Amsterdam, 1978. (9) Scouten, W. H. Affinity Chromatography: Bioselective Adsorption on Inert Matrices; Wiley: New York, 1981. (10) Chaiken, I. M., Ed.; Analytical Affinity Chromatography; CRC Press: Boca Raton, FL, 1987. (11) Loun, B.; Hage, D. S. Anal. Chem. 1996, 68, 1218. (12) Snyder, L. R.; Kirkland, J. J. Introduction to Modern Liquid Chromatography; Wiley: New York, 1979. (13) Unger, K. K., Ed.; Packings and Stationary Phases in Chromatographic Techniques; Marcel Dekker: New York, 1990. (14) Peters, E. C.; Svec, F.; Frechet, J. M. J. Adv. Mater. 1999, 11, 1169. (15) Wang, Q. C.; Svec, F.; Frechet, J. M. J. J. Chromatogr. 1994, 669, 230. (16) Minakuchi, H.; Nakanishi, K.; Soga, N.; Ishizuka, N.; Tanaka, N. J. Chromatogr., A 1997, 762, 135. (17) Wang, Q. C.; Svec, F.; Frechet, J. M. J. Anal. Chem. 1993, 65, 2243. (18) Ishizuka, N.; Minakuchi, H.; Nakanishi, K.; Soga, N.; Tanaka, N. J. Chromatogr., A 1998, 828, 83. (19) Luo, Q.; Zou, H.; Zhang, Q.; Xiao, X.; Ni, J. Biotech. Bioeng. 2002, 80, 481. (20) Pan, Z.; Zou, H.; Weimin, H.; Huang, X.; Wu, R. Anal. Chim. Acta 2002, 466, 141. (21) Petro, M.; Svec, F.; Frechet, J. M. J. Biotechnol. Bioeng. 1996, 49, 355. (22) Turkova, J.; Blaha, K.; Malanikova, M.; Vancurova, D.; Svec, F.; Kalal, J. Biochim. Biophys. Acta 1978, 524, 162. (23) Zemanova, I.; Turkova, J.; Capka, M.; Nakhapetyan, L. A.; Svec, F.; Kalal, J. Enzyme Microb. Technol. 1981, 3, 229. (24) Kucerova, Z.; Vankova, H.; Lenfeld, J.; Benes, M. J.; Turkova, J. Int. J. BioChromatogr. 1999, 4, 243. (25) Drobnik, J.; Saudek, V.; Svec, F.; Kalal, J.; Vojtisek, V.; Barta, M. Biotechnol. Bioeng. 1979, 21, 1317.
Analytical Chemistry, Vol. 76, No. 23, December 1, 2004 7013
form, which provides low nonspecific binding for biological compounds. It is also possible to use these diol groups for ligand attachment by adapting current approaches available for other affinity supports.28 This study will examine the development of high-performance affinity monolithic columns that contain immobilized human serum albumin (HSA). HSA is the most abundant protein in blood and is known to bind a variety of drugs and biological compounds.29-31 When immobilized in an HPAC column, this can be used under isocratic conditions as a weak-to-moderate strength ligand to study drug-protein interactions32 and separate some chiral solutes.11 This work will examine the immobilization of HSA onto GMA/ EDMA supports through four approaches: the epoxy method,33 the Schiff base method,34-40 the carbonyldiimidazole (CDI) method,41-45 and the disuccinimidyl carbonate (DSC) method.46 Each of the resulting supports will be evaluated in terms of its total protein content and its retention for several model compounds, including (R/S)-warfarin and D/L-tryptophan. The resulting data will provide information on the performance that can be expected from such columns and on the immobilization conditions that work best for the attachment of HSA and other proteins to GMA/EDMA monoliths. EXPERIMENTAL SECTION Reagents. The GMA (97% pure), EDMA (98% pure), 2,2′azobisisobutyronitrile (AIBN, 98% pure), 1-dodecanol (98% pure), N,N′-disuccinimidyl carbonate (>85% pure), and D-tryptophan (>99% pure) were from Aldrich (Milwaukee, WI). The cyclohexanol (>99% pure) was from Fluka (Milwaukee, WI). The 1,1′carbonyldiimidazole and HSA (Cohn fraction V, essentially fatty acid-free), racemic warfarin (3-(R-acetonylbenzyl)-4-hydroxycoumarin, >98% pure), racemic tryptophan, L-tryptophan (>98% pure), periodic acid or sodium periodate (>99% pure, an oxidizing agent), sodium borohydride (98% pure, a strong reducing agent), sodium (26) Drobnik, J.; Vlasak, J.; Pilar, J.; Svec, F.; Kalal, J. Enzyme Microb. Technol. 1979, 1, 107. (27) Hannibal-Friedrich, O.; Chun, M.; Sernetz, M. Biotechnol. Bioeng. 1980, 22, 157. (28) Kim, H. S.; Hage, D. S. In Handbook of Affinity Chromatography; Hage, D. S., Ed.; Marcel Dekker: New York; Chapter 3; in press. (29) Peters, T., Jr. All About Albumin: Biochemistry, Genetics and Medical Applications; Academic Press: San Diego, CA, 1995; Chapter 3. (30) Kragh-Hansen, U. Pharmacol. Rev. 1981, 33, 17. (31) Sengupta, A.; Hage, D. S. Anal. Chem. 1999, 71, 3821. (32) Yang, J.; Hage, D. S. J. Chromatogr., A 1997, 766, 15. (33) Berruex, L. G.; Freitag, R.; Tennikova, T. B. J. Pharm. Biomed. Anal. 2000, 24, 95. (34) Suda, Y.; Nakamura, M.; Koshida, S.; Kusumoto, S.; Sobel, M. J. Bioact. Compat. Polym. 2000, 15, 468. (35) Suzuki, N.; Quesenberry, M. S.; Wang, J. K.; Lee, R. T.; Kobayashi, K.; Lee, Y. C. Anal. Biochem. 1997, 247, 412. (36) Bjoerklund, M.; Hearn, M. T. W. J. Chromatogr., A 1996, 728, 149. (37) Frey, T.; Cosio, E. G.; Ebel, J. Phytochemistry 1993, 32, 543. (38) Thomas, D. H.; Beck-Westermeyer, M.; Hage, D. S. Anal. Chem. 1994, 66, 3823. (39) Stults, N. L.; Asta, L. M.; Lee, Y. C. Anal. Biochem. 1989, 180, 114. (40) Hornsey, V. S.; Prowse, C. V.; Pepper, D. S. J. Immunol. Methods 1986, 93, 83. (41) Koyama, T.; Terauchi, K. J. Chromatogr., B 1996, 679, 31. (42) Burton, S. J.; Stead, C. V.; Lowe, C. R. J. Chromatogr. 1990, 508, 109. (43) Taylor, R. F. Anal. Chim. Acta 1985, 172, 241. (44) Sudi, P.; Dala, E.; Szajani, B. Appl. Biochem. Biotechnol. 1989, 22, 31. (45) Potempa, L. A.; Motie, M.; Anderson, B.; Klein, E.; Baurmeister, U. Clin. Mater. 1992, 11, 105. (46) Hermanson, G. T. Bioconjugate Techniques; Academic Press: San Diego, 1996; Chapter 15.
7014
Analytical Chemistry, Vol. 76, No. 23, December 1, 2004
cyanoborohydride (94% pure, a mild reducing agent), 4-(dimethylamino)pyridine (DMAP), and tris(hydroxymethyl) aminomethane (99.8% pure) were from Sigma (St. Louis, MO). Reagents for the bicinchoninic acid (BCA) protein assay were from Pierce (Rockford, IL). The separate chiral forms of (R)- and (S)-warfarin (>98% pure) were obtained from DuPont Pharmaceuticals (Wilmington, DE). The acetic acid (>99.7% pure) was from EMD chemicals (Gibbstown, NJ). All aqueous solutions were prepared using water from a Nanopure system (Barnstead, Dubuque, IA) and filtered using Osmonics 0.22-µm nylon filters from Fisher (Pittsburgh, PA). Apparatus. The monoliths were prepared in 4.6-mm-i.d. × 5 cm stainless steel columns with PEEK inner liners from Alltech (Deerfield, IL), which included a special frit/insert that could be used to compress the monoliths and avoid gaps within the columns. Activating reagents for the monolithic columns were applied using a CM3000 pump from Milton Roy (Ivyland, PA). Immobilization of HSA was accomplished by passing protein solutions through the monoliths using a Waters 501 pump from Millipore (Milford, MA). The BCA protein assay was performed using a Shimadzu UV-160A spectrophotometer (Kyoto, Japan). The system used in the chromatographic studies consisted of a P4000 gradient pump and a UV100 absorbance detector from Thermoseparations (Riviera Beach, FL). Samples were injected using a Rheodyne Lab Pro valve (Cotati, CA) and a 20-µL sample loop. Chromatographic data were collected and processed using in-house programs written in Lab View 5.1 (National Instruments, Austin, TX). Preparation of Monolithic Columns. The general scheme used for preparing the monolithic columns and immobilizing HSA in these columns is shown in Figure 1. This scheme began with polymerizing the monolith, followed by washing with acetonitrile for 2 h at 0.5 mL/min. If required, this material was converted to a diol form by filling the column with a 0.5 M sulfuric acid solution and heating it, as described later for the Schiff base method. In the next step, the monolithic column was either used directly or activated by passing through a suitable activating agent. A 5 mg/mL HSA solution was then passed through this column for immobilization under the pH and buffer conditions required for the given coupling technique. The overall reaction used for monolith polymerization is shown in Figure 2. This involved the use of a 50:50 (v/v) mixture of GMA (the functional monomer) and EDMA (the cross linking agent) (Note: appropriate precautions should be used with GMA, which is highly toxic and a suspected carcinogen). This solution was combined with a 86:14 (v/v) mixture of cyclohexanol and 1-dodecanol, which were used as the porogens. The ratio of the GMA/ EDMA solution with the porogen solution was 40:60 (v/v). AIBN (a toxic agent and flammable solid) was used to initiate polymerization by being placed into this mixture at a 1% (w/w) level versus the monomer. After combining all of these chemicals, the polymerization mixture was sonicated for 10 min. Argon gas was passed through this mixture for another 10 min. Using a 1-mL syringe, the polymerization mixture was placed into an empty 4.6mm-i.d. × 5-cm column, with one side of this column then being sealed with a column plug. After this column had been filled, its other end was also sealed with a plug, and the column was placed in a water bath for 24 h at 60°C. After polymerization, the column was assembled in a suitable column housing. This involved the
Figure 1. General approach for the preparation of immobilized HSA monolithic columns.
Figure 2. Reaction for the copolymerization of glycidyl methacrylate with ethylene dimethacrylate. This figure shows only a portion of the polymer structure.
use of a special frit to allow slight compression of the polymer, thus reducing the effects of any shrinkage that occurred during the curing process. Any remaining porogens were then removed from the column by washing with 100 mL of acetonitrile at 0.5 mL/min for ∼3 h. Schiff Base Method. The reaction scheme used in the Schiff base method is shown in Figure 3a. A similar reaction involving the oxidation of a macroporous GMA/EDMA copolymer by periodic acid was reported by Svec and co-workers.47 This
began with a form of the monolith in which the original epoxy groups had been hydrolyzed to form diols. These diol groups were then oxidized by periodic acid to give aldehyde groups, which could react with primary amine groups on HSA (or other proteins) to form a Schiff base. This Schiff base was stabilized by using sodium cyanoborohydride to reduce it to a secondary amine. Any aldehyde groups that remained on the support after immobilization were reduced to alcohols by adding sodium borohydride. The conditions used for the Schiff base method were adapted from those described for diol-bonded silica.48 In this method, 5 mL of 0.5 M sulfuric acid was first passed through the epoxycontaining monolithic support. This material was converted into a diol form by heating it at 60°C for 4 h and washing with 100 mL of water at 0.5 mL/min (Note: this is the same technique used later to form diol monoliths for the CDI and DSC methods). Next, a 40-mL solution containing 2 g of periodic acid in a 90:10 (v/v) mixture of acetic acid and water was circulated through the diol monolith for 4 h at room temperature and 0.5 mL/min. A 20-mL solution containing 5 mg/mL HSA in pH 6.0, 1.5 M potassium phosphate buffer, also containing 100 mg of sodium cyanoborohydride, was then circulated through the column at 0.5 mL/min for 6 days at room temperature. A 20-mL portion of a 2.5 mg/mL sodium borohydride solution in pH 8.0, 0.1 M potassium phosphate buffer was then passed through the column at room temperature for 2 h at 0.5 mL/min to reduce any remaining aldehyde groups. The HSA column was washed for 4 h with pH 7.4, 0.067 M potassium phosphate buffer at 0.5 mL/min, with the column being stored in this same buffer at 4 °C until use. The (47) Svec, F.; Hrudkova, H.; Kalal, J. Angew. Makromol. Chem. 1978, 70, 101. (48) Larsson, P. O. Methods Enzymol. 1984, 104, 212.
Analytical Chemistry, Vol. 76, No. 23, December 1, 2004
7015
Figure 3. Methods used for the immobilization of HSA to a GMA/EDMA monolith.
resulting column showed good stability, with no significant changes in its retention being observed over at least three months. CDI Method. The reaction scheme for the CDI method is given in Figure 3b. All reactions were performed in a hood. This process also began by converting the epoxy groups in the monolith into diols. These diol groups were then reacted with 1,1′-carbonyldiimidazole to produce imidazolyl carbamate groups. A nucleophilic substitution reaction between these activated groups and primary amines on HSA (or another protein) was then used to form a stable amide linkage for immobilization. A diol-containing monolith was prepared in the same manner as described for the Schiff base method. After this reaction, the column was washed consecutively with 100 mL of water and 150 mL of acetonitrile in a hood, each applied at 0.5 mL/min. A solution containing 2 g of 1,1′-carbonyldiimidazole in 40 mL of acetonitrile was then passed through this column at room temperature for 12 h at 0.5 mL/min, followed by a wash with 150 mL of acetonitrile at the same flow rate. A 10-mL solution of 5 mg/mL HSA in pH 6.0, 1.5 M potassium phosphate buffer36 was next passed through the monolithic column at 0.5 mL/min for 20 7016
Analytical Chemistry, Vol. 76, No. 23, December 1, 2004
min, followed by another 10-mL portion of the same solution, applied at 0.5 mL/min and room temperature for 6 days. The column was later washed for 4 h with pH 7.4, 0.067 M potassium phosphate buffer at 0.5 mL/min and stored at 4°C in the same buffer prior to use. This column also gave good stability and showed no significant changes in its retention for the test analytes over three months. Epoxy Method. The reaction scheme for the epoxy method is given in Figure 3c. This involved the nucleophilic attack of an epoxy group on the monolith by amine groups on HSA (or other protein) under basic conditions, leading to the formation of a stable secondary amine linkage. In this method, a 5 mg/mL solution of HSA was prepared in pH 8.0, 1.5 M potassium phosphate buffer, and 20 mL of this solution was circulated through the monolithic column at 0.5 mL/min for 6 days at room temperature. After immobilization, any remaining epoxy groups were blocked by passing through the column a 60-mL portion of pH 8.0, 0.2 M Tris buffer49,50 at 0.5 mL/min for 2 h. The column was then washed with 120 mL of pH 7.4, 0.067 M potassium phosphate buffer at 0.5 mL/min for 4 h and stored at 4 °C in this buffer prior to use.
This column showed good stability during this study, with no significant changes in retention being noted over at least 3 months. DSC Method. The reaction scheme for the DSC immobilization method is illustrated in Figure 3d. In this technique, epoxy groups on the original monolith were first converted into diol groups under acidic conditions. These diol groups were then reacted with DSC to place succinimidyl carbonate groups on the monolith’s surface. These groups then reacted with primary amines on HSA to couple this protein to the monolith through a carbamate linkage. The conditions used in this work for the DSC method were similar to those described in ref 46 for the coupling of proteins with poly(ethylene glycol). A diol monolith was first prepared from an epoxy monolith as described for the Schiff base method. The column was then washed in a hood by applying 100 mL of water at 0.5 mL/min and 150 mL of acetonitrile at 0.5 mL/min for 4 h. A solution containing 100 mg of N,N′-disuccinimidyl carbonate in 20 mL of acetonitrile plus 49.2 mg of (dimethylamino)pyridine (DMAP) was then passed through the column in a hood at 0.5 mL/min for 12 h at room temperature. The column was washed with 150 mL of acetonitrile in a hood at 0.5 mL/min for 5 h to remove excess reagents. A 10-mL solution containing 5 mg/mL HSA in pH 7.5, 1.5 M potassium phosphate buffer was passed through the column at 0.5 mL/min for 20 min at room temperature. Another 10-mL solution of 5 mg/mL HSA in pH 7.5, 1.5 M potassium phosphate buffer was circulated through the column at 0.5 mL/min for 8 h at room temperature. Finally, the column was washed with 120 mL of pH 7.4, 0.067 M potassium phosphate buffer at 0.5 mL/min for 4 h and stored at 4°C in this buffer until use. Like the other supports used in this report, this column had good stability and gave no significant change in retention over three months. Evaluation of Monoliths. The amount of immobilized protein in each HSA monolith was determined by a BCA assay.51 For this assay, a representative column for each monolith was washed with 100 mL of water for 3 h at 0.5 mL/min. The monolith was then taken out of its column housing and ground into a fine powder. This powder was dried under vacuum overnight at room temperature. These samples and their corresponding blanks (i.e., the same monolithic material with no HSA present) were assayed in triplicate by the BCA method, using HSA as the standard. The mobile phase for all the chromatographic studies was pH 7.4, 0.067 M potassium phosphate buffer, which was degassed under vacuum for at least 30 min prior to use. All chromatographic studies were performed at 23 (( 1) °C. Fresh samples were prepared daily in the mobile phase and contained 20 µM of the desired analyte. Two or more 20-µL injections were made of each sample. No appreciable changes in retention times (i.e., random variations of
