Anal. Chem. 2002, 74, 4933-4936
Concentration of 17β-Estradiol Using an Immunoaffinity Porous Hollow-Fiber Membrane S. Nishiyama,† A. Goto,† K. Saito,*,†,‡ K. Sugita,† M. Tamada,§ T. Sugo,§ T. Funami,⊥ Y. Goda,⊥ and S. Fujimoto⊥
Department of Materials Technology, Chiba University, 1-33 Yayoi, Inage-ku, Chiba 263-8522, Japan, “Form and Function”, PRESTO, Japan Science and Technology Corporation, 1-18-9 Midori, Inage-ku, Chiba 263-0023, Japan, Takasaki Radiation Chemistry Research Establishment, Japan Atomic Energy Research Institute, 1233 Watanuki, Takasaki, Gunma 370-1292, Japan, Takeda Chemical Industry Ltd., 2-17-85 Juso-Honmachi, Yodogawa-ku, Osaka 532-8686, Japan
We describe attempts to achieve high throughput of 17βestradiol (E2) analysis, including the development of an immunocleanup membrane using polyclonal antibodies and an enzyme-linked immunosorbent assay (ELISA) using monoclonal antibodies. An epoxy-group-containing monomer, glycidyl methacrylate (GMA), was graft-polymerized onto a porous hollow-fiber membrane. Subsequently, anti-estrogen (ES) antibody, as a ligand, was coupled with the epoxy group. The ligand density ranged from 3.1 to 5.8 mg/g of the GMA-grafted porous hollowfiber membrane. A 1.0 µg/L E2 solution was forced to permeate through pores rimmed by the anti-ES-antibodyimmobilized polymer chains, at a constant permeation rate. A breakthrough curve, that is, the change in the E2 concentration of the effluent penetrating the outside of the hollow fiber with a change of the effluent volume, was determined. Bound E2 in amounts ranging from 0.42 to 0.80 µg was quantitatively eluted with 3-5 mL of methanol in the permeation mode. The higher permeation rate of the E2 solution resulted in the higher overall binding rate of E2 to the anti-ES-antibody-immobilized porous hollow-fiber membrane because of the negligible diffusional mass-transfer resistance of E2 to the antibody. Endocrine disruptors, with dioxins and organotins as representatives, significantly affect ecological systems.1-3 Gas chromatography/mass spectrometry (GC/MS) and enzyme-linked immunosorbent assay (ELISA) are powerful methods to determine the concentrations of endocrine disruptors in the environment; however, the cleanup or preconcentration of target molecules in environmental waters is essential. * Phone/fax: +81-43-290-3439. E-mail:
[email protected]. † Chiba University. ‡ Japan Science and Technology Corporation. § Japan Atomic Energy Research Institute. ⊥ Takeda Chemical Industry. (1) Horiguchi, H.; Takiguchi, N.; Cho, H. S.; Kojima, M.; Kaya, M.; Shiraishi, H.; Morita, M.; Hirose, H.; Shimizu, M. Mar. Environ. Res. 2000, 50, 223229. (2) Depledge, M. H.; Billinghurst, Z. Mar. Pollut. Bull. 1999, 39, 32-38. (3) Birnbaum, L. S. Toxicol. Lett. 1995, 82/83, 743-750. 10.1021/ac020141e CCC: $22.00 Published on Web 08/29/2002
© 2002 American Chemical Society
The cleanup of endocrine disruptors followed by analysis using a kit based on ELISA 4-7 must satisfy the following requirements: (1) selective isolation of the target molecule from congeners,that is, the interfering compounds present in complex matrixes; (2) decrease in the amount of organic solvents required for the cleanup; and (3) short sample cleanup time. Immunoaffinity purification using immunoaffinity beads is the only cleanup procedure that is specifically based on antigen-antibody interactions and that fulfills the first requirement. However, an immunosorbent should be developed to satisfy the latter two requirements. We have thus far prepared functional porous hollow-fiber membranes.8-10 Porous hollow-fiber membranes containing chelate-forming11,12 and ion-exchange13-16 groups have been prepared by radiation-induced graft polymerization and chemical modification. Heavy metal ions and proteins were adsorbed onto the chelating and ion-exchange porous hollow-fiber membranes during the permeation of the heavy metal ion and protein solutions across the membranes, respectively, with a negligible diffusional mass-transfer resistance. This is because the metal ions and proteins were transported by convective flow through the pores to minimize their diffusional path to the functional groups. (4) Petrovic, M.; Eljarrat, E.; de Alda, M. J. L.; Barcelo, D. Trends Anal. Chem. 2001, 20, 637-648. (5) Huang, C. H.; Sedlak D. L. Environ. Toxicol. Chem. 2001, 20, 133-139. (6) Goda, Y.; Kobayashi, A.; Fukuda, K.; Fujimoto, S.; Ike, M.; Fujita, M. Water Sci. Technol. 2000, 42, 81-88. (7) Fujita, M.; Ike, M.; Goda, Y.; Fujimoto, S.; Toyoda, Y.; Miyagawa, K. Environ. Sci. Technol. 1998, 32, 1143-1146. (8) Nakamura, M.; Kiyohara, S.; Saito, K.; Sugita, K.; Sugo, T. Anal. Chem. 1995, 71, 1323-1325. (9) Saito, K.; Tsuneda, S.; Kim, M.; Kubota, N.; Sugita, K.; Sugo, T. Radiat. Phys. Chem. 1999, 54, 517-525. (10) Saito, K. Sep. Sci. Technol. 2002, 37, 535-554. (11) Yamagishi, H.; Saito, K.; Furusaki, S.; Sugo, T.; Ishigaki, I. Ind. Eng. Chem. Res. 1991, 30, 2234-2237. (12) Konishi, S.; Saito, K.; Furusaki, S.; Sugo, T. Ind. Eng. Chem. Res. 1992, 31, 2722-2727. (13) Matoba, S.; Tsuneda, S.; Saito, K.; Sugo, T. Bio/Technology 1995, 13, 795797. (14) Tsuneda, S.; Saito, K.; Furusaki, S.; Sugo, T. J. Chromatogr. A 1995, 689, 211-218. (15) Tsuneda, S.; Shinano, H.; Saito, K.; Furusaki, S.; Sugo, T. Biotechnol. Prog. 1994, 10, 76-81. (16) Kubota, N.; Miura, S.; Saito, K.; Sugita, K.; Watanabe K.; Sugo, T. J. Membr. Sci. 1996, 117, 135-142.
Analytical Chemistry, Vol. 74, No. 19, October 1, 2002 4933
Figure 1. Preparation scheme of anti-ES-antibody-immobilized porous hollow-fiber membrane.
17β-Estradiol (E2) is a strong estrogenic compound that causes reproductive disturbances in animals.17-20 The E2 concentration in environmental waters ranges from 3 to 50 ng/L.21 In this study, we selected E2 as a model molecule of endocrine disruptors. The objective of this study was 2-fold: (1) to prepare an antiestrogen-antibody-immoblized porous hollow-fiber membrane, and (2) to demonstrate high-speed binding of spiked E2 samples and quantitative elution with methanol, which is ideally suited for simple high-throughput testing. To our knowledge, this is the first attempt to develop an immunoaffinity-interaction-based procedure for the analysis of E2, followed by ELISA. EXPERIMENTAL SECTION Preparation of Immunoaffinity Porous Membranes. A porous hollow-fiber membrane made of polyethylene was supplied by Asahi Kasei Corporation, Japan, and used as a trunk polymer for grafting. The inner and outer diameters of the hollow fiber were 1.8 and 3.1 mm, respectively. The hollow fiber had a threedimensional pore network with a porosity of 70% and pore diameter of 0.4 µm. Glycidyl methacrylate (GMA, CH2dCCH3COOCH2CHOCH2) was acquired from Tokyo Chemical Co. and was used without further purification. 17β-Estradiol (E2) was supplied by Takeda Chemical Industry. Anti-estrogen (ES) polyclonal antibody in immunized rabbit serum was purified with 40% ammonium sulfate precipitation and subsequent IgG fractionation with DEAE anion-exchange chromatography. Other reagents were of analytical grade or higher. An immunoaffinity porous hollow-fiber membrane was prepared by the following three steps (Figure 1): (1) Irradiation of an electron beam onto the trunk polymer to form radicals. The 10-cm-long porous hollow-fiber membrane was irradiated in a nitrogen atmosphere at ambient temperature using a cascade-type accelerator (Dynamitron IEA 3000-25-2, Radiation Dynamics Inc., New York). The dose was 200 kGy. (2) Grafting of an epoxy-groupcontaining vinyl monomer (GMA). The irradiated hollow fiber was immersed in a 10%(v/v) GMA/methanol solution. After 12 min, the hollow fiber was taken out and washed thoroughly to remove (17) Crain, D. A.; Guillette, L. J., Jr.; Rooney, A. A.; Pickford, D. B. Environ. Health Persp. 1997, 105, 528-533. (18) Wiese, T. E.; Kelce, W. R. Chem. Ind. 1997, 16, 648-653. (19) Janssen, P. A. H.; Lambert, J. G. D.; Vethaak, A. D.; Goos, H. J. Th. Aquat. Toxicol. 1997, 39, 195-214. (20) Eric, I. F.; Perrin, C. W. Pediatrics 2000, 105, e55-e57. (21) Desbrrow, C.; Routledge, E. J.; Brighty, G. C.; Sumpter, J. P.; Waldock, M. Environ. Sci. Technol. 1998, 32, 1549-1558.
4934
Analytical Chemistry, Vol. 74, No. 19, October 1, 2002
residual GMA and poly-GMA homopolymer. The degree of GMA grafting is defined as
dg (%) ) 100(weight increase due to grafting)/ (weight of the trunk polymer) (1) Here, dg was set at 170%. The resultant GMA-grafted porous hollow-fiber membrane was referred to as the GMA fiber. (3) Coupling of antibody to some epoxy groups. The GMA fiber was immersed in a 1.75 mg/mL anti-ES antibody dissolved in 0.025 M borate buffer (pH 10) for a prescribed coupling time ranging from 4 to 36 h at 277 K. After coupling, the fiber was washed with the borate buffer. The resultant porous hollow-fiber membrane containing anti-ES antibody was referred to as ES-IA(q) fiber, where q designates the ligand density, that is, the amount of anti-ES antibody coupled. Binding of E2 to Immunoaffinity Porous Hollow-Fiber Membranes. The ES-IA(q) fiber of 2.0 cm effective length was positioned as shown in Figure 2. Then, 1.0 µg/L of E2 in 10%(v/ v) methanol/water was permeated through the pores from the inside surface of the hollow fiber to the outside surface at a constant permeation rate ranging from 30 to 300 mL/h at ambient temperature. The E2 concentration of the effluent penetrating the outside surface of the ES-IA(q) fiber was continuously determined using an ES ELISA kit (Takeda Chemical Industry). The E2 concentration changes in the effluent were plotted against the effluent volume to obtain the breakthrough curves. The equilibrium binding capacity (EBC) of E2 was calculated as
EBC (µg/g) )
∫
Ve
0
(C0 - C) dV/W
(2)
where C0 and C are the E2 concentrations of the feed and effluent, respectively. V, Ve, and W are the effluent volume, the effluent volume when C reached C0, and the weight of the GMA fiber, respectively. The molar binding ratio of E2 to anti-ES antibody under the assumption that one anti-ES antibody of the IgG type can bind two E2 molecules is expressed as
molar binding ratio of E2 to anti-ES antibody immobilized ) EBC/[2(amount of anti-ES antibody immobilized)] (3) Subsequently, the pores of the immunoaffinity porous hollowfiber membrane were washed with 10%(v/v) methanol/water at
Figure 2. Experimental apparatus for E2 recovery using ES-IA fiber.
Figure 3. Amount of anti-ES antibody immobilized vs coupling time.
Figure 4. Breakthrough curves of E2 for various amounts of antiES antibody immobilized.
a permeation rate corresponding to that in the E2 binding procedure at ambient temperature. A 3-5-mL portion of methanol was permeated across the hollow-fiber membrane to elute E2 bound to the ES-IA(q) fiber. The E2 concentration of the effluent penetrating the outside surface was continuously determined using the ES ELISA kit. The elution percentage was defined by dividing the amount of eluted E2 by the amount of bound E2. The adsorption, washing, and elution procedures were repeated twice. The amount of bound E2 and the elution percentage were determined after each cycle.
RESULTS AND DISCUSSION The ligand density, that is, the amount of anti-ES antibody immobilized, is shown in Figure 3 as a function of reaction time at 277 K in the coupling reaction of anti-ES antibody with the epoxy group of the poly-GMA chain grafted onto a porous hollow-fiber membrane. The ligand density increased with increasing coupling time. The ligand density for a coupling time of 36 h amounted to 5.8 mg/g of the GMA fiber. The binding and elution of E2 using an ES-IA(q) fiber, that is, an anti-ES-antibody-immobilized porous hollow-fiber membrane, were performed in the permeation mode. Breakthrough curves of E2 for the ES-IA(q) fiber with various q’s ranging from 3.1 to 5.8 mg/g are shown in Figure 4 at a constant permeation rate of 60 mL/h at ambient temperature. The equilibrium binding capacity (EBC) of E2, evaluated using eq 2, is shown in Figure 5,
Figure 5. Equilibrium binding capacity of E2 vs amount of immobilized anti-ES antibody.
along with the molar binding ratio calculated using eq 3. The EBC of E2 increased linearly with increasing ligand density; the molar binding ratio was constant at 0.043. The possible reasons for the observed low molar binding ratio are as follows: (1) deterioration of the antibody molecules due to the change in their threedimensional structures during the coupling procedure, (2) unfavorable orientation of the immobilized antibody for the antigen as a result of the coupling, and (3) lower specificity of the antibody to the antigen with a lower molecular mass than to proteins. Analytical Chemistry, Vol. 74, No. 19, October 1, 2002
4935
Figure 6. Dependence of breakthrough curves on permeation rate of E2 solution through the pores of ES-IA(4.8) fiber.
Figure 7. Elution percentage as a function of amount of anti-ES antibody immobilized onto the porous hollow-fiber membrane.
The E2 solution was permeated through pores rimmed by the anti-ES-antibody-immobilized polymer chains at various solution permeation rates of 30-300 mL/h. The residence time of the solution through the pores, tr, can be calculated as
tr ) π(do2 - di2)L/[4 (permeation rate)]
(4)
Figure 8. Equilibrium binding capacity of E2 for the ES-IA(4.4) fiber vs cycle number.
diameters, and the length of the ES-IA(q) fiber, respectively. The breakthrough curves overlapped, regardless of the residence time of the E2 solution, ranging from 13 to 1.3 s, as shown in Figure 6; a higher permeation rate of the E2 solution resulted in a higher overall binding rate of E2 to the anti-ES-antibody-immobilized porous hollow-fiber membrane. This is because the diffusional mass-transfer resistance of E2 to the immobilized antibody was negligible, which is favorable for high-throughput cleanup of E2. Various amounts of the E2 bound to the 2.0-cm-long ES-IA(q) fiber were eluted with methanol. The elution percentage was almost 100% irrespective of the ligand density, as shown in Figure 7. Quantitative elution with common and safe organic solvents, one of the requirements of the cleanup of E2, was achieved. The amount of E2 bound to the ES-IA(4.4) fiber decreased with an increasing number of cycles, as shown in Figure 8. The amount of bound E2 at the second cycle was 40% of that at the first cycle. Methanol as an eluate may induce the conformational change of anti-ES antibody immobilized onto the porous hollowfiber membrane, resulting in reduction of its binding capacity for E2. ACKNOWLEDGMENT We thank Kohei Watanabe and Noboru Kubota of Asahi Kasei Corporation, Japan, for providing the starting porous hollow-fiber membrane. Received for review March 5, 2002. Accepted July 7, 2002.
where , do, di, and L are the porosity, the outer and inner
4936 Analytical Chemistry, Vol. 74, No. 19, October 1, 2002
AC020141E
