Phenylboronic AcidSalicylhydroxamic Acid Bioconjugates. 2

240

Bioconjugate Chem. 2001, 12, 240−250

Phenylboronic Acid-Salicylhydroxamic Acid Bioconjugates. 2. Polyvalent Immobilization of Protein Ligands for Affinity Chromatography Jean P. Wiley, Karin A. Hughes, Robert J. Kaiser, Edward A. Kesicki,† Kevin P. Lund, and Mark L. Stolowitz* Prolinx, Inc., 22322 20th Avenue SE, Bothell, Washington 98021. Received August 8, 2000; Revised Manuscript Received January 11, 2001

Phenylboronic acid bioconjugates prepared from alkaline phosphatase by reaction with either 2,5dioxopyrrolidinyl 3-{N-[3-(1,3,2-dioxaboran-2-yl)phenyl]carbamoyl}propanoate (PBA-XX-NHS) or 2,5dioxopyrrolidinyl 6-{[3,5-di-(1,3,2-dioxaboran-2-yl)phenyl]carbonylamino}hexanoate (PDBA-X-NHS) were compared with respect to the efficiency with which they were immobilized on salicylhydroxamic acid-modified Sepharose (SHA-X-Sepharose) by boronic acid complex formation. When immobilized on moderate capacity SHA-X-Sepharose (5.4 µmol of SHA/mL of gel), PDBA-alkaline phosphatase conjugates were shown to be stable with respect to both the alkaline (pH 11.0) and acidic (pH 2.5) buffers utilized to recover anti-alkaline phosphatase during affinity chromatography. Boronic acid complex formation was compared to covalent immobilization of alkaline phosphatase on Affi-Gel 10 and Affi-Gel 15. PDBA-AP‚SHA-X-Sepharose was shown to afford superior performance to both AffiGel 10 and Affi-Gel 15 with respect to immobilization of alkaline phosphatase, retention of antialkaline phosphatase and recovery of anti-alkaline phosphatase under alkaline conditions. High capacity SHA-X-Sepharose (g7 µmol of SHA/mL of gel) was shown to afford superior performance to moderate capacity SHA-X-Sepharose (4.5 µmol of SHA/mL of gel) with respect to stability at pH 11.0 and pH 2.5 when a PDBA-RHuman IgG conjugate with a low incorporation ratio of only 1.5:1 was immobilized on SHA-X-Sepharose and subsequently utilized for affinity chromatography of Human IgG. The results are interpreted in terms of either a bivalent or trivalent interaction involving boronic acid complex formation.

INTRODUCTION

In the previous paper (1) we introduced a small molecule polyvalent system that exploits the interaction between phenylboronic acid (PBA) and salicylhydroxamic acid (SHA) for protein immobilization. The utility of this system was demonstrated by preparing a PBA-modified alkaline phosphatase conjugate (PBA-AP), immobilizing the conjugate on a SHA-modified chromatographic support (SHA-Sepharose), and utilizing the resulting chromatographic support for affinity chromatography of antialkaline phosphatase. Initial results were promising, although a potential limitation was identified relating to the need to append several PBA moieties when preparing conjugates. The present paper describes the further optimization of the PBA‚SHA system with respect to the immobilization of protein ligands for affinity chromatography. Our efforts were directed toward optimizing the reagent components of the system so as to minimize the requisite number of phenylboronic acid moieties while concomitantly increasing both the efficiency and stability with which PBA conjugates are immobilized on SHA chromatographic supports. We report here the development of a new boronic acid reagent that incorporates two equivalents of boronic acid per site of conjugation, PDBA-X-NHS, as well as the development of a new high capacity chromatographic * To whom correspondence should be addressed. Fax: (425) 487-9578. E-mail: [email protected]. † Present address: ICOS Corporation, 22021 20th Avenue SE, Bothell, WA 98021.

support, SHA-X-Sepharose. The performance of PDBAX-NHS and SHA-X-Sepharose is compared to that of PBA-XX-NHS and SHA-Sepharose, respectively. In nature, polyvalent interactions involving the simultaneous binding of multiple ligands on one biological entity (a macromolecule or cell surface) to multiple receptors on another are often collectively much stronger than the corresponding monovalent interactions. A recent review of polyvalent systems states, “The possibility that multiple simultaneous interactions have unique collective properties that are qualitatively different from properties displayed by their constituents, which interact monovalently, suggests new strategies for the design of drugs and research reagents for biochemistry and biology” (2). Although polyvalent systems usually involve protein receptors, dimeric derivatives prepared from vancomycin and peptides containing D-Ala-D-Ala have recently provided small molecule model systems for investigating polyvalency (3-5). Trimeric derivatives prepared from vancomycin and L-Lys-D-Ala-D-Ala were reported to bind to one another 25 times tighter than avidin‚biotin, one of the strongest biological interactions known (6). MATERIALS AND METHODS

Materials. Reagents used in the syntheses were purchased from Aldrich (Milwaukee, WI) and Fluka (Ronkonkoma, NY) and utilized without further purification. 2,5-Dioxopyrrolidinyl 6-(3-{N-[3-(1,3,2-dioxaboran2-yl)phenyl]carbamoyl}propanoylamino)hexanoate (PBAXX-NHS), methyl 4-aminomethylsalicylate hydrochloride, and SHA-Sepharose were prepared as previously re-

10.1021/bc000086l CCC: $20.00 © 2001 American Chemical Society Published on Web 03/01/2001

Polyvalent Immobilization of Protein Ligands

ported (1). 2,5-Dioxopyrrolidinyl 6-{[3,5-di(1,3,2-dioxaboran-2-yl)phenyl]carbonylamino}hexanoate (PDBA-XNHS) and SHA-X-Sepharose were prepared as described below and are now available from Prolinx (Bothell, WA). EAH Sepharose 4B and Sephadex G25 were obtained from Amersham Pharmacia Biotech (Piscataway, NJ). Affi-Gel 10 and Affi-Gel 15 were obtained from Bio-Rad (Hercules, CA). Alkaline phosphatase (EC 3.1.3.1) was obtained from Calzyme Laboratories (San Luis Obispo, CA) and rabbit anti-alkaline phosphatase (IgG fraction) was obtained from Rockland (Gilbertsville, PA). Goat anti-human IgG and ChromPure proteins (purified whole molecule immunoglobulins from normal Human serum) were obtained from Jackson (West Grove, PA). 1H, 11B, and 13C NMR spectra were recorded at 300, 121, and 75 MHz, respectively, on a Varian Gemini 2000 spectrometer with a broadband probe. Internal referencing was employed and chemical shifts are reported as δ values. Coupling constants are reported in hertz. Highresolution mass spectra of synthetic intermediates and reagents were recorded on a PE Biosystems Mariner ESITOF mass spectrometer in the positive ion mode. The purity of synthetic products was determined by analytical HPLC on a Hewlett-Packard 1050Ti system equipped with a diode array detector. A 2.1 cm × 0.22 cm C4 column (Aquapore Butyl, PE Biosystems) equilibrated with 0.1 M triethylammonium acetate, pH 6.5, was eluted at a flow rate of 0.5 mL/min after an initial period of 5 min with a linear gradient of from 0 to 100% methanol over 30 min. Product purity exceeded 95% in every instance. Protein bioconjugates were characterized with respect-to-incorporation ratio on a Hewlett-Packard 8453 diode array UV-vis spectrophotometer. 3,5-Di(dihydroxyboryl)benzoic Acid (1). A total of 5.0 g (17.9 mmol) of 3,5-dibromobenzoic acid was dissolved in 300 mL of THF and cooled to -78 °C. A total of 12.7 mL (17.8 mmol) of methyllithium was added over 2 min and stirred for 5 min, and 58.0 mL (98.6 mmol) tertbutyllithium was added over 5 min. The solution was stirred at -78 °C for 15 min, and the solution was allowed to warm to -45 °C, stirred for 45 min, and then again cooled to -78 °C. A total of 27.0 mL (117 mmol) of triisopropylborate was added, stirred for 15 min at -78 °C, and then allowed to warm to room temperature. The solution was stirred at room temperature for 2 h, and the bulk of solvent was removed by rotary evaporation and residual solvent under high vacuum while warming for 1 h to 60 °C. The solids were dissolved in 60 mL of water, cooled to 0 °C for 1 h and then filtered off solids (borate salts). The filtrate was cooled to 0 °C and adjusted pH to ∼4 by dropwise addition of concentrated HCl. The solution was allowed to stand overnight at 4 °C, and solids were collected by filtration, washed with cold water, and dried under high vacuum. The volume of filtrate was reduced to ∼70 mL, and again the solution was allowed to stand at 4 °C overnight. Finally, solids were collected by filtration, washed with cold water, and dried under high vacuum to give 2.23 g (60%) of 1 as a white powder. 1H NMR (DMSO-d6): δ 8.18 (s, 4H), 8.41 (s, 3H), 12.65 (br s, 1H). 13C NMR (DMSO-d6): δ 129.1, 133.5, 137.0, 144.9, 168.4. 11B NMR (DMSO-d6): δ 27.7. HRMS (ES+) calcd for C7H8B2O6Na: 233.0405. Obsd: 233.0410. 6-{[3,5-Di(dihydroxyboryl)phenyl]carbonylamino}hexanoic Acid (3). A total of 3.2 g (15.2 mmol) of 1 was suspended in 300 mL of 1,4-dioxane, 2.2 mL (30.4 mmol) of 1,3-propanediol and heated to near reflux while stirring until the solution became homogeneous. The solvent was removed by rotary evaporation, the solid was

Bioconjugate Chem., Vol. 12, No. 2, 2001 241

dissolved by heating in 150 mL of 1,4-dioxane. Again, the solvent was removed by rotary evaporation, the solid was dissolved by heating in 150 mL of 1,4-dioxane, and then the solvent was removed by rotary evaporation. The solid was dissolved with heating in 400 mL of 1,4-dioxane, the solution was allowed cool to room temperature, and then 1.84 g (16.0 mmol) of N-hydroxysuccinimide was added followed by 3.30 g (16.0 mmol) of DCC. The solution was stirred gently overnight at room temperature, during which time a white precipitate formed, the precipitate was filtered from solution, and the filtrate was concentrated by rotary evaporation. The intermediate was dissolved (2, colorless solid) in 100 mL of methanol, 3.99 g (30.4 mmol) of 6-aminohexanoic acid (3.99 g, 30.4 mmol) and 10.6 mL (60.9 mmol) of diisopropylethylamine were added. The solution was stirred for 24 h at room temperature, the solvent was concentrated by rotary evaporation, and then the product was dissolved in 75 mL of 0.25 N NaOH. The solid was filtered off, and the filtrate cooled to 0 °C and then acidified to pH ∼4 with 1 N HCl. The product was allowed to stand for 1 day at 4 °C. Finally, the precipitate was collected by filtration, washed with cold water, and dried under high vacuum to give 3.59 g (73%) of 3 as a white solid. 1H NMR (DMSO-d6): δ 1.29 (m, 2H), 1.51 (m, 4H), 2.19 (t, J ) 7.3 Hz, 2H), 3.22 (q, J ) 6.3 Hz, 2H), 8.05 (br s, 4H), 8.19 (s, 2H), 8.28 (s, 1H), 8.47 (t, J ) 11.1 Hz, 1H), 11.92 (br s, 1H). 13C NMR (CD OD): δ 24.0, 25.7, 28.3, 33.1, 39.0, 132.7, 3 133.3, 141.5, 169.5, 176.1. HRMS (ES+) calcd for C13H19B2NO7Na: 346.1245. Obsd: 346.1254. 2,5-Dioxopyrrolidinyl 6-{[3,5-Di(1,3,2-dioxaboran2-yl)phenyl]carbonylamino}hexanoate (4, PDBA-X -NHS). A total of 5.67 g (17.6 mmol) of 3 was suspended in 200 mL of 1,4-dioxane, 2.55 mL (35.3 mmol) of 1,3propanediol and heated stirred suspension for 15 min under reflux. Decanted solution into another flask leaving behind an oily material. The solvent was removed by rotary evaporation, the solid was dissolved in 150 mL of 1,4-dioxane. Again, the solvent was removed by rotary evaporation, the solid was dissolved in 150 mL of 1,4dioxane and the solvent was removed by rotary evaporation. The solid was dissolved in 200 mL of 1,4-dioxane, 2.18 g (18.9 mmol) of N-hydroxysuccinimide and 3.80 g (18.4 mmol) of DCC and stirred at room temperature overnight, during which time a white precipitate formed. The precipitate was filtered from solution, and the filtrate was concentrated by rotary evaporation. The product was dissolved in a minimum volume of 1,4-dioxane and added to 250 mL of rapidly stirred methyl tert-butyl ether. The product initially precipitates as an oil but affords a white solid after 24 h of rapid stirring. The product was collected by filtration and washed with methyl tert-butyl ether to give 6.52 g (74%) of 4 as a white powder. 1H NMR (DMSO-d6): δ 1.38 (m, 2H), 1.53 (m, 2H), 1.64 (m, 2H), 1.99 (p, J ) 5.1 Hz, 4H), 2.66 (t, J ) 7.3 Hz, 2H), 2.79 (s, 4H), 3.22 (q, J ) 6.4 Hz, 2H), 4.10 (t, J ) 5.4 Hz, 8H), 8.13 (s, 1H), 8.16 (s, 2H), 8.49 (t, J ) 5.5 Hz, 1H,). 13C NMR (DMSO-d6): δ 24.0, 25.4, 25.5, 26.9, 28.6, 30.1, 38.9, 61.6, 131.8, 133.4, 134.5, 141.6, 166.7, 169.2, 170.5. HRMS (ES+) calcd for C17H22B2N2O9Na (4 minus two 1,3propanediols): 443.1409. Obsd: 443.1429. Methyl 4-{[(tert-Butoxy)carbonylamino]methyl}salicylate (5). A total of 10.9 g (50 mmol) of methyl 4-aminomethylsalicylate hydrochloride was suspended in 200 mL of anhydrous methanol, and 8.7 g (50 mmol) of di-tert-butyl dicarbonate and 7.0 mL (50 mmol) of Et3N

242 Bioconjugate Chem., Vol. 12, No. 2, 2001

were added. The solid rapidly dissolved with the slow evolution of gas. The reaction mixture was stirred under dry N2 for 18 h at room temperature and then evaporated to dryness to afford an amorphous white solid. The solid was partitioned between 200 mL of ethyl acetate and 100 mL of water, the layers were separated, ethyl acetate was dried over Na2SO4, the solution was filtered, and the filtrate was evaporated to a white solid. The solid was crystallized from 1:1 ethyl acetate:hexanes, and the product was filtered and dried in vacuo to give 13.7 g (97%) of 5 as a white solid, mp: 95-96 °C (uncorrected). 1H NMR (CDCl ): δ 1.42 (s, 9H), 3.90 (s, 3H), 4.26 (d, J 3 ) 6 Hz, 2H), 4.99 (t, J ) 6 Hz, 1H), 6.75 (d, J ) 8 Hz, 1H), 6.84 (s, 1H), 7.73 (d, J ) 8 Hz, 1H), 10.72 (s, 1H). 13C NMR (CDCl ): δ 28.5, 44.4, 52.4, 80.0, 111.5, 115.9, 3 118.2, 130.5, 150.0, 156.3, 162.1, 170.8. HRMS (ES+) calcd for C14H19NO5Na: 304.1161. Obsd: 304.1152. 4-{[(tert-Butoxy)carbonylamino]methyl}salicylic Acid (6). A total of 8.7 g (30.9 mmol) of 5 was dissolved in 100 mL of dry THF, and 4.4 g (30.9 mmol) of potassium trimethylsilanolate (90% pure) was added and heated for 24 h under reflux, during which time a tan precipitate formed. The mixture was evaporated to dryness, and the solid was dissolved in 100 mL of water. The stirred solution was cooled in an ice-water bath, and the pH was adjusted to ∼2 by dropwise addition of saturated aqueous KHSO4, thereby producing an offwhite precipitate. The precipitate was filtered, washed with ice-cold water, and dried in vacuo over KOH pellets to give 6.7 g (81%) of 6 as a white solid, mp: 141-144 °C (uncorrected, decomposes with effervescence). 1H NMR (CDCl3): δ 1.47 (s, 9H), 4.33 (d, J ) 6 Hz, 2H), 5.07 (t, J ) 6 Hz, 1H), 6.80 (d, J ) 8 Hz, 1H), 6.88 (s, 1H), 7.81 (d, J ) 8 Hz, 1H), 10.72 (br s, 2H). 13C NMR (CDCl3): δ 28.5, 44.4, 80.5, 111.1, 115.9, 118.3, 131.5, 148.5, 156.6, 162.6, 174.1. HRMS (ES+) calcd for C13H17NO5Na: 290.1004. Obsd: 290.1005. Cyanomethyl 4-{[(tert-Butoxy)carbonylamino]methyl}salicylate (7). A total of 8.2 g (30.6 mmol) of 6 was suspended in 25 mL of chloroacetonitrile, 4.3 mL (30.6 mmol) of Et3N was added, and the mixture was stirred under dry N2 for 8 h at 50 °C. The mixure was cooled to room temperature, the solvent was evaporated, and the residue was partitioned between 250 mL of ethyl acetate and 250 mL of water. The layers were separated, and ethyl acetate was washed with 100 mL of saturated aqueous NaHCO3 and 100 mL of saturated aqueous NaCl. The solution was dried over Na2SO4 and filtered, and the filtrate was evaporated to dryness. The pale tan solid was dissolved in 100 mL of ethyl acetate, and the produce was adsorbed onto 10 g of silica gel. Silica gel on a glass frit was collected by vacuum filtration and washed with 100 mL of ethyl acetate, and the filtrate was evaporated to dryness. The residue was crystallized from 1:1 ethyl acetate:hexanes to give 7.9 g (88%) of 7 as an off-white solid, mp: 144-146 °C (uncorrected). 1H NMR (CDCl3): δ 1.45 (s, 9H), 4.30 (d, J ) 6 Hz, 2H), 5.00 (s, 2H), 5.05 (t, J ) 6 Hz, 1H), 6.83 (d, J ) 8 Hz, 1H), 6.91 (s, 1H), 7.77 (d, J ) 8 Hz, 1H), 10.12 (s, 1H). 13C NMR (CDCl ): δ 28.4, 44.3, 49.0, 80.1, 109.6, 114.2, 3 116.1, 118.7, 130.5, 149.7, 156.2, 162.5, 168.5. HRMS (ES+) calcd for C15H18N2O5Na 329.1113. Obsd: 329.1111. Cyanomethyl 4-(Aminomethyl)salicylate Hydrochloride (8). A total of 7.7 g (26.2 mmol) of 7 was dissolved in 150 mL of THF, and dry HCl gas was slowly bubbled through the solution. After 5 min, the gas was discontinued , and the reaction was stirred for 30 min at room temperature, during which time a creamy white precipitate formed. The reaction mixture was cooled for

Wiley et al.

2 h in an ice-water bath. The precipitate was filtered, washed with 100 mL of diethyl ether, and dried in vacuo over KOH pellets to give 5.8 g (91%) of 8 as a white powder. 1H NMR (DMSO-d6): δ 4.00 (s, 2H), 5.20 (s, 2H), 7.05 (d, J ) 8 Hz, 1H), 7.15 (s, 1H), 7.75 (d, J ) 8 Hz, 1H), 8.62 (br s, 3H), 10.38 (s, 1H). 13C NMR (DMSO-d6): δ 41.6, 49.7, 113.0, 116.1, 118.0, 119.7, 131.1, 142.3, 159.5, 166.1. HRMS (ES+) calcd for C10H11N2O3: 207.0770. Obsd: 207.0768. 4-[N-({4-[(Cyanomethyl)oxycarbonyl]-3-hydroxyphenyl}methyl)carbamoyl]butanoic Acid (9). A total of 1.22 g (5.0 mmol) of 8 was suspended in 100 mL dry DCM and stirred in an ice-water bath, and a solution of 0.57 g (5.0 mmol) of glutaric anhydride and 0.7 mL (5.0 mmol) of Et3N was added in 25 mL of dry DCM over 15 min. The reaction was stirred for 18 h at room temperature. The mixture was evaporated to dryness, and the solid was triturated under 50 mL of ice-cold 0.1 N HCl. The solid was collected by filtration, washed with cold water, and dried in vacuo over KOH pellets to give 1.43 g (89%) of 9 as a white solid, mp: 125-126 °C (uncorrected). 1H NMR (DMSO-d6): δ 1.75 (q, J ) 7 Hz, 2H), 2.19 (t, J ) 7 Hz, 2H), 2.22 (t, J ) 7 Hz, 2H), 4.24 (d, J ) 6 Hz, 2H), 5.17 (s, 2H), 6.81 (d, J ) 8 Hz, 1H), 6.85 (s, 1H), 7.70 (d, J ) 8 Hz, 1H), 8.39 (t, J ) 6 Hz, 1H), 10.5 (br s, 1H), 11.7 (br s, 1H). 13C NMR (DMSOd6): δ 20.7, 33.1, 34.5, 41.8, 49.7, 111.2, 115.9, 116.2, 118.5, 130.9, 149.1, 160.2, 166.6, 172.3, 174.5. HRMS (ES+) calcd for C15H16N2O6Na: 343.0906. Obsd: 343.0909. 2,5-Dioxopyrrolidinyl 4-[N-({4-[(Cyanomethyl)oxycarbonyl]-3-hydroxyphenyl}methyl)-carbamoyl]butanoate (10). A total of 1.00 g (3.1 mmol) of 9 was dissolved in 50 mL of dry THF, 0.36 g (3.1 mmol) of N-hydroxysuccinimide and 0.64 g (3.1 mmol) of DCC were added, and the mixture was gently stirred under dry N2 for 1 h at room temperature, during which time a white precipitate formed. The reaction was stirred for 24 h at room temperature and then cooled for 4 h at -20 °C. The precipitate was filtered from cold solution and washed with 10 mL of THF. The filtrate and wash were combined and evaporated to dryness. The residue was crystallized from 1:1 2-propanol:hexanes and filtered and dried in vacuo to give 1.04 g (80%) of 10 as white needles, mp 116-119 °C (uncorrected). 1H NMR (DMSO-d6): δ 1.88 (q, J ) 7 Hz, 2H), 2.30 (t, J ) 7 Hz, 2H), 2.71 (t, J ) 7 Hz, 2H), 2.80 (s, 4H), 4.26 (d, J ) 6 Hz, 2H), 5.18 (s, 2H), 6.82 (d, J ) 8 Hz, 1H), 6.86 (s, 1H), 7.71 (d, J ) 8 Hz, 1H), 8.44 (t, J ) 6 Hz, 1H), 10.18 (br s, 1H). 13C NMR (DMSO-d6): δ 20.4, 25.5, 29.7, 33.6, 41.8, 49.6, 111.2, 115.9, 116.2, 118.4, 130.9, 148.9, 160.1, 166.5, 169.0, 170.5, 171.7. HRMS (ES+) calcd for C19H19N3O8Na: 440.1070. Obsd: 440.1068. Preparation of SHA-X-Sepharose. Collected EAH Sepharose 4B (50 mL slurry) by vacuum filtration in a Buchner funnel fitted with filter paper (Whatman #1) and washed with 100 mL of deionized water followed by 100 mL of 0.1 M NaHCO3 buffer, pH 9.8. The moist cake was suspended in 12.5 mL of 0.1 M NaHCO3 buffer, pH 9.8, and transferred into a 50 mL wide-mouth screw cap plastic bottle. A total of 1.252 g (3.0 mmol) of 10 was dissolved in 3 mL of dry DMF and added to the reaction bottle. The bottle was rotated for 1 h at room temperature. Sepharose was collected on a Buchner funnel and washed with 200 mL of deionized water. The moist cake was transferred to a 200 mL wide-mouth screw cap plastic bottle. A total of 8.686 g (0.125 mmol) of hydroxylamine hydrochloride in 12.5 mL of 1 M NaHCO3, pH 10.0, was added to 100 mL of deionized water. The pH of hydroxylamine solution was adjusted to 10.0 by


dropwise addition of 10 N NaOH, and then the final volume of the solution was adjusted to 125 mL by adding deionized water. A total of 125 mL of freshly prepared hydroxylamine solution was added to the reaction bottle and the bottle rotated for 16 h at room temperature. Finally, SHA-X-Sepharose was collected on a Buchner funnel, washed with 250 mL of deionized water, and stored as a slurry in 20% (v/v) ethanol/water. Determined SHA concentration of modified Sepharose by first suspending 1 mL of gel in 2 mL of 50% (v/v) glycerol/water, removing a 0.5 mL aliquot of the suspension, and further diluting with 2.5 mL of 50% (v/v) glycerol/water. Finally, the SHA concentration was estimated by comparing the absorbance of the suspension of SHA-modified gel at 320 nm (320 ) 4900) to that of unmodified EAH Sepharose 4B. Enzymatic Activity of PBA- and PDBA-AP Conjugates. Alkaline phosphatase (AP, 50 000 units; Sigma) was dialyzed in a Slide-A-Lyzer (3500 MWCO, Pierce) against 2 L of 0.1 M NaHCO3, pH 8.3, for 48 h at 4 °C. Determined concentration of AP dialysate by measuring the absorbance at 280 nm (280 ) 107 500). Conjugated six aliquots of AP dialysate (∼4.3 mg/mL) by adding a 5-, 10-, or 50-fold molar excess of either PBA-XX-NHS (100 mM stock solution in DMF) or PDBA-X-NHS (100 mM stock solution in DMF). Reactions were incubated on ice for 1.5 h, and conjugates were transferred to SlideA-Lyzers (3500 MWCO, Pierce) and dialyzed against 2 L of 0.1 M NaHCO3, pH 8.3, at 4 °C over 48 h. Incorporation ratios (to within 0.5:1) were estimated by comparing the ratio of the absorbance at 260 nm (260 ) 4500 for PBA and 260 ) 4000 for PDBA) to the absorbance at 280 nm. Enzymatic activity was assayed with 4-nitrophenyl phosphate (pNPP) as substrate. A stock solution of pNPP was prepared by dissolving 15 mg of pNPP in 15 mL of DEA buffer (1 M diethanolamine, 1 mM MgCl2, 0.1 mM ZnSO4, pH 10.2). Diluted 5 aliquots (100 µL) of each of the six conjugates to 0.05 µg/mL with TBS buffer (10 mM tris(hydroxymethyl)aminomethane hydrochloride, 140 mM NaCl, pH 8.0), in a 96-well plate, and assayed for AP activity by adding 100 µL of pNPP stock solution to each well and incubating for 10 min at 37 °C. Enzymatic activity was determined of conjugates relative to unmodified AP from the average absorbance at 405 nm obtained from each of five aliquots. A405 values were determined on a Bio-Tek Instruments ELX808 plate reader. Immobilization of PBA- and PDBA-AP Conjugates on SHA-X-Sepharose. Each of six SHA-XSepharose columns (0.5 mL in 0.5 × 15 cm column; 5.4 µmol of SHA/mL of gel) were equilibrated with 50 mL of 0.1 M NaHCO3, pH 8.3. Aliquots of 0.5 mL of each of the conjugates prepared above were diluted to 5 mL total volume with 0.1 M NaHCO3, pH 8.3. Loaded conjugates on SHA-X-Sepharose columns by simple gravity-flow, and washed columns with 10 mL of 0.1 M NaHCO3, pH 8.3 (a sufficient quantity of buffer for the A280 to return to baseline). Determined the quantity of conjugate immobilized on each column by subtracting the quantity of conjugate in each of the eluents (as determined by the absorbance at 280 nm) from the quantity of conjugate loaded onto each column. Affinity Chromatography of RAP on PBAAP‚SHA-X-Sepharose and PDBA-AP‚SHA-XSepharose. Reconstituted 25 mg of rabbit anti-alkaline phosphatase Ab (RAP, Rockland) in 5 mL of water and dialyzed in a Slide-A-Lyzer (3500 MWCO; Pierce) against 2 L of 0.1 M NaHCO3, pH 8.3, for 48 h at 4 °C. Aliquots of 4 mg of RAP were diluted with 0.1 M NaHCO3, pH 8.3, to a final volume of 5 mL and loaded by simple


gravity-flow onto each of the six columns prepared as described above containing 0.5 mL of either PBA-AP‚ SHA-X-Sepharose or PDBA-AP‚SHA-X-Sepharose. Columns were washed with 15 mL of 0.1 M NaHCO3, pH 8.3, collecting the first 10 mL fraction of each eluent. Determined protein concentration in each of the fractions from the absorbance at 280 nm (280 ) 210 000) and calculated the quantity of RAP retained on each column by subtracting the quantity of RAP in each of the fractions from the quantity loaded onto each column. Initially, RAP was eluted from each column with 15 mL of 50 mM phosphate buffer, pH 11.0 (high pH eluent), collecting a 10 mL fraction into a tube containing 1 mL of 1 M phosphate buffer, pH 5.75. Subsequently, additional RAP was eluted from each column with 15 mL of 100 mM glycine hydrochloride buffer, pH 2.5 (low pH eluent), collecting a 10 mL fraction. Low pH of the eluent fraction was adjusted to pH 8-9 by addition of 1 drop of 10 N NaOH. Determined protein concentration in each of the fractions from the absorbance at 280 nm and calculated recovery of RAP. High pH eluent fractions and low pH eluent fractions for affinity purified RAP were analyzed by SDS-PAGE on 7.5% gels according to Laemmli (7). Aliquots of 25 µL of each fraction were diluted with 25 µL of reducing application buffer [63 mM tris(hydroxymethyl)minomethane hydrochloride, pH 6.8, 10% (v/v) glycerol, 2% (w/v) SDS; 0.05% (v/v) 2-mercaptoethanol, and 0.001% (w/v) bromophenol blue] and heated for 3 min at 95 °C prior to loading 10 µL/lane. PAGE was run at 200 V until bromophenol blue reached the bottom of the plate. Visualized proteins by staining gel with Silver stain (BioRad). Immobilization of AP on SHA-X-Sepharose, Affi-Gel 10, and Affi-Gel 15. Six alkaline phosphatase samples (300 000 units each, Sigma) were dialyzed in six Slide-A-Lyzers (3500 MWCO; Pierce) against 2 L of 0.1 M NaHCO3, pH 8.3, over 48 h at 4 °C. The concentration of each AP dialysate was determined from the absorbance at 280 nm (280 ) 107 500). A total of 2.0 mg of AP dialysate was conjugated by adding a 15-fold molar excess of PDBA-X-NHS (100 mM stock solution in DMF). The reaction was incubated on ice for 1.5 h, and then the conjugate was dialyzed in a Slide-A-Lyzer (3500 MWCO; Pierce) against 2 L of 0.1 M NaHCO3, pH 8.3, over 24 h at 4 °C. The extent of PDBA incorporation was determined from the ratio of the absorbance at 260 nm (260 ) 4000) to the absorbance at 280 nm. A SHA-X-Sepharose column (0.25 mL in 0.5 × 15 cm column; 5.3 µmol of SHA/ mL of gel) was equilibrated with 50 mL of 0.1 M NaHCO3, pH 8.3. PDBA-AP conjugate was loaded on the column by simple gravity-flow, the column was washed with 50 mL of 0.1 M NaHCO3, pH 8.3, and the first 10 mL fraction of eluent was collected. Protein concentration in the eluent was determined from the absorbance at 280 nm (280 ) 107 500), and the quantity of PDBA-AP immobilized on the column was calculated by subtracting the quantity of PDBA-AP in the eluent from the quantity loaded onto the column. A total of 0.25 mL of Affi-Gel 10 (Bio-Rad) and 0.25 mL of Affi-Gel 15 (Bio-Rad) were transferred to each of two 1.5 mL micro test tubes. Gels were washed three times each by vortexing with 1 mL of cold deionized water, centrifuging at low speed to pellet the gel, and then removing the supernatant. A total of 2.0 mg of AP dialysate was added to each of the gel suspensions. Rotated Affi-Gel 10 suspension for 2 h at room temperature and Affi-Gel 15 suspension for 4 h at 4 °C. A total of 0.25 mL of a 1 M solution of glycine ethyl ester was


added, pH 8.0, to each of the gel suspensions to consume excess reactive sites and rotated for 1 h at room temperature. Gel suspensions were transferred to 0.5 × 15 cm columns and washed with 50 mL of 0.1 M NaHCO3, pH 8.3, collecting the first 10 mL fraction from each column. Determined the protein concentrations of the two fractions from the absorbance at 280 nm (280 ) 107 500), and calculated the quantity of AP immobilized on each of the columns by subtracting the quantity of AP in each of the fractions from the quantity loaded onto each of the columns. Affinity Chromatography of RAP on PDBAAP•SHA-X-Sepharose, AP-Affi-Gel 10, and AP-AffiGel 15. A total of 50 mg of rabbit anti-alkaline phosphatase Ab (RAP, Rockland) was reconstituted in 5 mL of deionized water and dialyzed in a Slide-A-Lyzer against 2 L of 0.1 M NaHCO3, pH 8.3, for 48 h at 4 °C. Aliquots of 2.0 mg of RAP dialysate were diluted with 0.1 M NaHCO3, pH 8.3, to a final volume of 5 mL and loaded by simple gravity-flow onto each of the three 0.5 × 15 cm columns prepared as described above, containing 0.25 mL of either PDBA-AP‚SHA-X-Sepharose (5.3 µmol of SHA/mL of gel), AP-Affi-Gel 10, or AP-Affi-Gel 15. The columns were eluted by gravity-flow sequentially applying 15 mL of 0.1 M NaHCO3, pH 8.3, 15 mL of 50 mM sodium phosphate buffer, pH 11.0 (high pH eluent) and 15 mL of 100 mM glycine hydrochloride buffer, pH 2.5 (low pH eluent). The first 10 mL fraction of the high pH eluent and low pH eluent was collected from each of the columns. The protein concentration of each of the fractions was determined from the absorbance at 280 nm (280 ) 210000) and calculated the recovery of RAP associated with each fraction by subtracting the quantity of protein in each fraction from the quantity loaded onto the corresponding column. Analyzed high pH fractions and low pH fractions for affinity purified RAP by SDS-PAGE as described above. Purification of Human IgG on PDBA-RHuman IgG‚SHA-X-Sepharose. A total of 10.0 mg of goat antihuman IgG (RHuman IgG, Jackson) was dialyzed in a Slide-A-Lyzer (3500 MWCO; Pierce) overnight against 2 L of 0.1 M NaHCO3, pH 8.3, at 4 °C. The protein concentration of dialysate was determined from the absorbance at 280 nm (280 ) 210 000). A total of 8.0 mg of RHuman IgG dialysate was conjugated by adding a 10-fold molar excess of PDBA-X-NHS (100 mM stock solution in DMF). The reaction was incubated on ice for 1 h, and then the conjugate was desalted on a Sephadex G25 column (13 cm × 1.5 cm) eluted with 0.1 M NaHCO3, pH 8.3. The column effluent was monitored at 280 nm, and the PDBA-RHuman IgG fraction was collected. The extent of PDBA incorporation was determined from the ratio of the absorbance at 260 nm (260 ) 4000) to the absorbance at 280 nm. Approximately 1.5 mg of PDBARHuman IgG was diluted to a total volume of 5.0 mL with 0.1 M NaHCO3, pH 8.3, and loaded onto a SHA-XSepharose column (0.25 mL in 0.5 × 15 cm column; 7.1 µmol of SHA/mL of gel). The column was washed with 10 mL of 100 mM glycine hydrochloride buffer, pH 2.5, to remove PDBA-RHuman IgG which was unstable with respect to the low pH eluent and then equilibrated column with 50 mL of 0.1 M NaHCO3, pH 8.3. A total of 25.0 mg of ChromPure proteins (Jackson) was reconstituted in 10 mL of 0.1 M NaHCO3, pH 8.3, and then loaded by simple gravity-flow onto a PDBA-RHuman IgG‚SHAX-Sepharose column. Human IgG was eluted with 15 mL of 50 mM phosphate buffer, pH 11.0, collecting a 10 mL fraction into a tube containing 1 mL of 1 M phosphate buffer, pH 5.75. Additional Human IgG was eluted with

Wiley et al.

15 mL of 100 mM glycine hydrochloride buffer, pH 2.5, collecting a 10 mL fraction. The pH of the low pH eluent fraction was adjusted to pH 8-9 by addition of 1 drop of 10 N NaOH. The protein concentration of each fraction was determined from the absorbance at 280 nm (280 ) 210 000), and the recovery of Human IgG was calculated by subtracting the quantity of protein in each fraction from the quantity loaded onto the column. Fractions for Human IgG were analyzed by SDSPAGE on 12.5% gels according to Laemmli (7). Aliquots of 25 µL of each fraction were diluted with 25 µL of reducing application buffer and heated for 3 min at 95 °C prior to loading 10 µL/lane. PAGE was run at 200 V until bromophenol blue reached the bottom of the plate. Proteins were transferred onto ProBlott membranes (Applied Biosystems) with a Mini Trans-Blot Cell (BioRad) by electroblotting at 50 V for 2 h at room temperature in 10 mM CAPS buffer, pH 11, containing 10% methanol. Nonspecific binding to ProBlott membranes was blocked by incubating overnight at room temperature in freshly prepared TBS buffer [10 mM tris(hydroxymethyl)aminomethane hydrochloride, 140 mM NaCl, pH 7.4), containing 5% (w/v) nonfat dried milk followed by washing three times for 5 min with TBST (TBS containing 0.5% (v/v) Tween 20]. Membranes were incubated with either monoclonal anti-human IgG-alkaline phosphatase conjugate (1:50000 dilution, Sigma) in TBST or rabbit anti-goat IgG-alkaline phosphatase conjugate (1:50000 dilution, Sigma) in TBST for 2 h at room temperature with constant agitation. Membranes were washed three times for 5 min with TBST and developed by incubating with 5-bromo-4-chloro-3-indolyl phosphate and nitroblue tetrazolium (BCIP/NBT FAST Tablets, Sigma). Finally, after sufficient color developed, membranes were rinsed briefly in methanol and dried. RESULTS AND DISCUSSION

The structures of the target reagents 2,5-dioxopyrrolidinyl 6-{[3,5-di(1,3,2-dioxaboran-2-yl)phenyl]carbonylamino}hexanoate (4, PDBA-X-NHS) and 2,5-dioxopyrrolidinyl 4-[N-({4-[(cyanomethyl)oxycarbonyl]-3-hydroxyphenyl}methyl)carbamoyl]butanoate (10, cyanomethyl SA-X-NHS) are illustrated in Figure 1. Also illustrated, for comparison, are the related reagents 2,5-dioxopyrrolidinyl 6-{3-[N-(3-(1,3,2-dioxaboran-2-yl)phenyl)carbamoyl]propanoylamino}hexanoate (PBA-XX-NHS) and methyl 4-[(6-aminohexanoylamino)methyl]salicylate hydrochloride (methyl SA-X-Amine) which were previously prepared (1). The structures of the chromatographic supports SHA-X-Sepharose (12) and SHA-Sepharose are illustrated in Figure 2. The results that follow compare the performance of the reagents and chromatographic supports described herein to those previously prepared with respect to the immobilization of protein ligands for affinity chromatography. Syntheses of Reagents. Reagent 4 was prepared by the synthesis outlined Scheme 1. The starting material, 3,5-dibromobenzoic acid, was sequentially reacted with methyllithium, tert-butyllithium, and triisopropylborate in THF to afford 3,5-di(dihydroxyboryl)benzoic acid (1) in 60% yield. Protection of the boronic acid moieties of 1 with 1,3-propanediol in 1,4-dioxane and subsequent reaction with N-hydroxysuccinimide and DCC next produced the intermediate NHS ester 2. Reaction of 2 with 6-aminohexanoic acid and diisopropylethylamine in methanol produced the long-chain carboxylic acid 3 in 73% yield. Finally, protection of the boronic acid moieties of 3 with 1,3-propanediol in 1,4-dioxane and subsequent



Scheme 1. Synthesis of 2,5-Dioxopyrrolidinyl 6-{[3,5-Di(1,3,2-dioxaboran-2-yl)phenyl]carbonylamino}hexanoate (4, PDBA-X-NHS)a

a (i) Methyllithium, tert-butyllithium, and triisopropylborate in THF; (ii) aqueous HCl (pH 95%) (>95%) (>95%)

1.42 1.42 1.22 1.40 1.37 1.03

0.93 (66%)f 0.68 (48%)f 0.38 (31%) 0.42 (30%) 0.38 (28%) 0.31 (30%)

0.43 (30%)f 0.27 (19%)f 0.11 (9%) 0.43 (31%) 0.15 (11%) 0.08 (8%)

1.36 (96%)f 0.95 (67%)f 0.49 (40%) 0.85 (61%) 0.53 (39%) 0.39 (38%)

a Incorporation ratio is defined as the ratio of moles of reagent incorporated per mole of protein modified. b AP conjugates (2.0 mg) in 5 mL of 100 mM sodium bicarbonate buffer, pH 8.3, were immobilized on 0.5 mL of SHA-X-Sepharose (5.5 µmol of SHA/mL of gel). c Crude RAP (5 mg) in 5 mL of 100 mM sodium bicarbonate buffer, pH 8.3, was applied to either PBA-APe‚SHA-X-Sepharose or PDBAAP‚SHA-X-Sepharose. d Recovered in 10 mL of 50 mM sodium phosphate buffer, pH 11.0. e Recovered in 10 mL of 100 mM glycine hydrochloride buffer, pH 2.5. f SDS-PAGE revealed the presence of contaminating PBA-AP in recovered RAP.

complexed with immobilized SHA. Alternatively, these results could be interpreted to imply at any time only a single boronic acid complex need be present to retain the immobilized species but that as many as six boronic acid moieties may to required to ensure that on average at least one boronic acid complex is always present. Affinity Chromatography of anti-Alkaline Phosphatase on PBA-AP‚SHA-X-Sepharose and PDBAAP‚SHA-X-Sepharose. Results obtained for purification

of RAP on PBA-AP‚SHA-X-Sepharose and PDBA-AP‚ SHA-X-Sepharose are also summarized in Table 2. In every instance, AP columns efficiently retained RAP at pH 8.3, which was subsequently recovered in either an alkaline (pH 11.0) or acidic (pH 2.5) eluent. However, the results involving PBA-AP‚SHA-X-Sepharose are misleading in that they represent an overestimate of from about 1 to 10% with respect to protein recovery owing to the presence of trace amounts of contaminating PBA-


Wiley et al.

Table 3. Affinity Chromatography of anti-Alkaline Phosphatase on PDBA-AP‚SHA-X-Sepharose, Alkaline Phosphatase Affi-Gel 10 and Alkaline Phosphatase Affi-Gel 15

Sepharosef Affi-Gel 10g Affi-Gel 15h

AP immobilized (mg)a

RAP retained (mg)b

pH 11 RAP recovered (mg)c

pH 2.5 RAP recovered (mg)d

total RAP recovered (mg)e

2.9 (73%) 1.5 (38%) 1.0 (25%)

5.2 (24%) 3.9 (18%) 1.6 (8%)

2.5 (48%) 0.66 (17%) 0.48 (30%)

0.72 (14%) 0.45 (12%) 0.28 (18%)

3.22 (62%) 1.11 (29%) 0.76 (48%)

a In each instance, 4.0 mg of AP was applied to 0.25 mL of gel. b In each instance, 21.6 mg of crude RAP was applied to the column. Recovered in 10 mL of 50 mM phosphate buffer, pH 11.0. d Recovered in 10 mL of 100 mM glycine hydrochloride buffer, pH 2.5. e Additional RAP can be recovered by increasing the ionic strength of the eluent by adding 200 mM NaCl. f PDBA-AP (8.7:1, 4.0 mg) was applied to a column of SHA-X-Sepharose (0.25 mL of 6.24 µmol of SHA/mL of gel). g AP reacted with Affi-Gel 10 for 2 h at room temperature. h AP reacted with Affi-Gel 15 for 4 h at 4 °C.

c

Table 4. Affinity Chromatography of Human IgG on PDBA-rHuman IgG‚SHA-X-Sepharose SHA in Sepharose (µmol/mL of gel)

PDBA-RHuman IgG retained (mg)a

protein retained (mg)b

protein recovered, pH 11.0 (mg)c

protein recovered, pH 2.5 (mg)d

total protein recovered (mg)

4.1 4.5 7.1

1.35 (90%) 1.41 (94%) 1.38 (92%)

1.35 1.37 0.86

0.23 (17%)f 0.19 (14%)f 0.26 (30%)

0.43 (32%)e 0.24 (18%)e 0.16 (19%)

0.66 (49%) 0.43 (31%) 0.42 (49%)

a PDBA-RHuman IgG (1.5 mg) was immobilized on SHA-X-Sepharose (0.25 mL). b Large excess of ChromPure proteins (purified whole molecule immunoglobulins from normal Human serum) applied so that the amount retained was proportional to the amount of immobilized PDBA-RHuman IgG. c Recovered in 10 mL of 50 mM sodium phosphate buffer, pH 11.0. d Recovered in 10 mL of 100 mM glycine hydrochloride buffer, pH 2.5. e PDBA-RHuman IgG detected in Western blot with enzyme-linked detection employing RGoat IgG-AP for detection of (Goat) PDBA-RHuman IgG. f Trace amounts ofPDBA-RHuman IgG detected in Western blot with enzyme-linked detection employing RGoat IgG-AP for detection of (goat) PDBA-RHuman IgG.

Figure 4. (A) SDS-PAGE of fractions containing RAP recovered from SHA-X-Sepharose in pH 11.0 eluent; (B) SDS-PAGE of fractions containing RAP recovered from SHA-X-Sepharose in pH 2.5 eluent; (1) RAP recovered from PBA-AP (4:1 incorporation ratio) column; (2) RAP recovered from PBA-AP (6:1 incorporation ratio) column; (3) RAP recovered from PBA-AP (18.5:1 incorporation ratio) column; (4) RAP recovered from PDBA-AP (3:1 incorporation ratio) column; (5) RAP recovered from PDBA-AP (4.5:1 incorporation ratio) column; (6) RAP recovered from PDBA-AP (11.5:1 incorporation ratio) column. Note that the arrows indicate the presence of contaminating PBA-AP in fractions recovered from PBA-AP‚SHA-X-Sepharose columns. The mobility of PBA and PDBA conjugates is related to the incorporation ratio with low incorporation ratios affording the least reduction in electrophoretic mobility and high incorporation ratios affording the greatest reduction in electrophoretic mobility.

AP that was revealed by SDS-Page. Figure 4 illustrates SDS-PAGE analyses of RAP fractions recovered from both PBA-AP‚SHA-X-Sepharose and PDBA-AP‚SHA-

X-Sepharose columns at pH 11.0 and 2.5. PBA-AP conjugates with average incorporation ratios of 4:1 and 6:1 were found to be present as contaminants in both high pH and low pH eluent fractions, indicating that some portion of the immobilized conjugate was unstable with respect to the elution conditions. On the other hand, SDS-PAGE revealed that both the high pH and low pH eluent fractions recovered from PDBA-AP‚SHA-XSepharose columns were essentially free of contaminating conjugate. Affinity Chromatography of anti-Alkaline Phosphatase on PDBA-AP‚SHA-X-Sepharose, AP-AffiGel 10, and AP-Affi-Gel 15. The performance of PDBAAP‚SHA-X-Sepharose was compared to that of AP-AffiGel 10 and AP-Affi-Gel 15 with respect to affinity chromatography of anti-alkaline phosphatase so as to compare the approach described herein with covalent immobilization of a protein ligand. Affi-Gel 10 and AffiGel 15 are activated immunoaffinity supports for coupling ligands with primary amino groups by reaction with support-bound N-hydroxysuccinimidyl esters (7). Similar to Sepharose in many respects, Affi-Gels are prepared from beaded cross-linked agarose and are reported by the manufacturer (Bio-Rad) to have 15 µmol of N-hydroxysuccinimidyl ester/mL of gel. Affi-Gel 10 contains a hydrophilic 10-atom spacer, whereas Affi-Gel 15 contains a hydrophilic 15-atom spacer with a central quaternary amine. Results for affinity chromatography of RAP on immobilized AP are summarized in Table. 3. PDBA-AP‚ SHA-X-Sepharose proved significantly more efficient than both AP-Affi-Gel 10 and AP-Affi-Gel 15 with respect to immobilization of alkaline phosphatase, retention of anti-alkaline phosphatase, and recovery of anti-alkaline phosphatase at high pH. PDBA-AP‚SHA-X-Sepharose proved comparable to both AP-Affi-Gel 10 and AP-AffiGel 15 with respect to recovery of anti-alkaline phosphatase at low pH. From a practical standpoint, immobilization of PDBA-AP on SHA-X-Sepharose proved far less laborious and time consuming (70 min, including preparation and desalting of PDBA-conjugate) than immobilization of AP on either Affi-Gel 10 (210 min) or AffiGel 15 (330 min).


Purification of Human IgG on PDBA-RHuman IgG‚SHA-X-Sepharose. The purification of Human IgG on PDBA-RHuman IgG‚SHA-X-Sepharose was employed to investigate the impact of immobilized SHA concentration on the performance of the PDBA‚SHA system. Goat anti-human IgG was conjugated with PDBA-X-NHS at an input ratio of 10:1 to afford a conjugate (PDBARHuman IgG) with an average incorporation ratio of only 1.5:1. PDBA-RHuman IgG was then immobilized on each of three SHA-X-Sepharose columns with SHA concentrations of 7.1, 4.5, and 4.1 µmol of SHA/mL of gel. The columns were washed with low pH eluent to remove any loosely complexed conjugate prior to equilibration. A protein fraction from normal human serum comprised primarily of whole molecule immunoglobulins was applied to each of the columns and Human IgG was recovered by eluting the columns with high pH eluent followed by low pH eluent. The results are summarized in Table 4. The fractions were reduced, separated by SDS-PAGE, blotted onto membranes, and analyzed by enzyme-linked immunological detection employing either monoclonal anti-human IgG-alkaline phosphatase conjugate (RHuman IgG-AP) for detection of human IgG or rabbit anti-goat IgG-alkaline phosphatase conjugate (RGoat IgG-AP) for detection of PDBA-RHuman IgG (Figure 5). The Western blot detected with anti-goat IgGAP confirmed the immunoreactivity of the conjugate with respect to PDBA-RHuman IgG and revealed the presence of trace amounts of PDBA-RHuman IgG in the low pH fractions obtained from each of the columns (although PDBA-RHuman IgG is barely visible in the low pH fraction recovered from the SHA-X-Sepharose column having 7.1 µmol of SHA/mL of gel). Qualitatively, the amount of contaminating conjugate detected appears to be inversely proportional to the SHA concentration of the column. Owing to the low incorporation ratio of the PDBA-RHuman IgG conjugate (only 1.5:1) it is somewhat remarkable that more contaminating conjugate was not detected. We concluded above that for efficient immobilization and subsequent stability with respect to both high pH and low pH eluents, either two or three of the boronic acid moieties associated with PDBA conjugates must be available for simultaneous complex formation with immobilized SHA. The results described immediately above imply that, in this instance, only two boronic acid moieties need be available for complex formation with immobilized SHA to confer stability when the effective SHA concentration is relatively high. Since the reported incorporation ratio (1.5:1) represents an average value (implying on average three boronic acid moieties per RHuman IgG), the contaminating conjugate found in the low pH fractions may have resulted from that fraction of the conjugates having only a single PDBA moiety. It is interesting to note that empirical studies involving the purification of recombinant C-terminal polyhistidine fusion proteins on immobilized metal ion affinity chromatography columns indicate that six is the requisite number of C-terminal histidine residues needed to routinely enable efficient retention of fusion proteins on columns charged with nickel ions. This observation is particularly remarkable in light of the fact that nickel ion complexation is know to involve the participation of only two adjacent histidine residues (12-16). CONCLUSIONS

Alkaline phosphatase conjugates prepared from either PDBA-X-NHS or PBA-XX-NHS were directly compared


Figure 5. (A) Western blot of Human IgG fractions obtained at high pH (11.0) and low pH (2.5) from PDBA-RHuman IgG‚ SHA-X-Sepharose columns (7.1, 4.1, and 4.5 µmol of SHA/mL of gel). Blot prepared to detect the presence of residual PDBARHuman IgG. Stained with anti-human IgG-alkaline phosphatase and BCIP/NBT substrate; (B) Western blot of Human IgG fractions obtained at high pH (11.0) and low pH (2.5) from PDBA-RHuman IgG‚SHA-X-Sepharose columns (7.1, 4.1, and 4.5 µmol of SHA/mL of gel). Blot prepared to detect the presence of residual PDBA-RHuman IgG. Stained with rabbit anti-goat IgG-alkaline phosphatase and BCIP/NBT substrate.

with respect to the efficiency with which they were immobilized on SHA-X-Sepharose by PBA‚SHA complex formation. For efficient immobilization, it was necessary to prepare conjugates having a minimum of either three appended PDBA moieties or six appended PBA moieties. When immobilized on high capacity SHA-X-Sepharose, alkaline phosphatase conjugates were shown to be stable with respect to both the alkaline (50 mM phosphate buffer, pH 11.0) and acidic (100 mM glycine HCl buffer, pH 2.5) buffers that were utilized to recover anti-alkaline phosphatase during affinity chromatography. From indirect evidence we conclude that for efficient immobilization and subsequent stability, on average either two or three boronic acid moieties associated with a protein conjugate must be available for complex formation with immobilized SHA. Boronic acid complex formation was compared to direct covalent immobilization of alkaline phosphatase on AffiGel 10 and Affi-Gel 15 for affinity chromatography of


anti-alkaline phosphatase. PDBA-AP‚SHA-X-Sepharose was shown to afford superior performance to both APAffi-Gel 10 and AP-Affi-Gel 15 with respect to immobilization of alkaline phosphatase, retention of antialkaline phosphatase, and recovery of anti-alkaline phosphatase at pH 11.0. Furthermore, immobilization of PDBA-AP on SHA-X-Sepharose proved both less laborious and less time-consuming than immobilization of alkaline phosphatase on either Affi-Gel 10 or Affi-Gel 15. High capacity SHA-X-Sepharose (g7 µmol of SHA/mL of gel) was shown to afford superior performance to moderate capacity SHA-X-Sepharose (4.1-4.5 µmol of SHA/mL of gel) with respect to immobilized ligand stability at pH 11.0 and pH 2.5 when a PDBA-RHuman IgG conjugate with an incorporation ratio of only 1.5:1 was immobilized. LITERATURE CITED (1) Stolowitz, M. L., Ahlem, C., Hughes, K. A., Kaiser, R. J., Kesicki, E. A., Li, G., Lund, K. P., and Wiley, J. P. (2000) Phenylboronic Acid-Salicylhydroxamic Acid Bioconjugates. 1. A Novel Boronic Acid Complex for Protein Immobilization. Bioconjugate Chem. 12, 229-239. (2) Mammen, M., Choi, S.-K., and Whitesides, G. M. (1998) Polyvalent Interactions in Biological Systems: Implications for Design and Use of Multivalent Ligands and Inhibitors. Angew. Chem., Int. Ed. 37, 2754-2794. (3) Rao, J., and Whitesides, G. M. (1997) Tight Binding of a Dimeric Derivative of Vancomycin with Dimeric L-Lys-D-AlaD-Ala. J. Am. Chem. Soc. 119, 10286-10290. (4) Rao, J., Yan, L., Lahiri, J., Whitesides, G. M., Weis, R. M., and Warren, H. S. (1998) Binding of a Dimeric Derivative of Vancomycin to L-Lys-D-Ala-D-lactate in Solution and at a Surface. Chem. Biol. 6, 353-359. (5) Rao, J., Yan, L., Xu, B., and Whitesides, G. M. (1999) Using Surface Plasmon Resonance to Study the Binding of Vancomycin and its Dimer to Self-Assembled Monolayers Presenting D-Ala-D-Ala. J. Am. Chem. Soc. 121, 2629-2630.

Wiley et al. (6) Rao, J., Lahiri, J., Issacs, L., Weis, R. M., and Whitesides, G. M. (1998) A Trivalent System from Vancomycin‚D-Ala-DAla with Higher Affinity than Avidin‚Biotin. Science 280, 708-711. (7) Laemmli, U. K. (1970) Cleavage of Structural Proteins During Assembly of the Head of Bacteriophage T4. Nature 227, 680-685. (8) Schwyzer, R., Iselin, B., and Feurer, M.(1955) U ¨ ber Aktivierte Ester I. Aktivierta Ester der Hippursaüre und ihre Umsetzungen mit Benzylamin. Helv. Chim. Acta 38, 69-79. (9) Schwyzer, R., Feurer, M. Iselin, B., and Kagi, H. (1955) U ¨ ber Aktivierte Ester II. Synthese Aktivierter Ester von Aminosaüre-derivaten. Helv. Chim. Acta 38, 80-83. (10) Schwyzer, R., Feurer, M., and Iselin, B. (1955) U ¨ ber Aktivierte Ester III. Umsetzungen Akktivierter Ester von Aminosaüre-und Peptid-derivaten mit Aminen und Aminosaüreestern. Helv. Chim. Acta 38, 83-91. (11) Iselin, B., Feurer, M., and Schwyzer, R. (1955) U ¨ ber Aktivierte Ester V. Verwendung der Cyanmethylester-Methode zur Herstellung von (N-Carbobenzoxy-S-benzyl-L-cysteinyl)-L-tyrosyl-L-isoleucin auf Verschiedenen Wegen. Helv. Chim. Acta 38, 1508-1516. (12) Schwyzer, R., Iselin, B., Rittel, W., and Sieber, P. (1956) U ¨ ber aktivierte ester VII. Synthesen Zyklischer Polypeptide. C-Tetraglycyl und C-Hexaglycyl. Helv. Chim. Acta 39, 872883. (13) Hochuli, E. (1989) Cost-Effective Purification of Bioproteins. Chem. Ind. 112, 69-70. (14) Hochuli, E. (1990) Purification of Recombinant Proteins with Metal Chelate Adsorbant. Genet. Eng. 12, 87-98. (15) Hochuli, E. (1992) Interrelations of Chemistry and Biotechnology III. Pure Appl. Chem. 64, 169-84. (16) Hochuli, E., and Piesecki, S. (1992) Interaction of Hexahistidine Fusion Proteins with Nitrilotriacetic Acid-Chelated Nickel(2+) Ions. Methods 4, 68-72. (17) Crowe, J., Doebeli, H., Gentz, R., Hochuli, E., Stueber, D., and Henco, K. (1994) 6×His-Ni-NTA Chromatography as a Superior Technique in Recombinant Protein Expression/ Purification. Methods Mol. Biol. 31, 371-87.

BC000086L

Phenylboronic AcidSalicylhydroxamic Acid Bioconjugates. 2

Recommend Documents