Anal. Chem. 2001, 73, 3392-3399
An Immunoassay for Small Analytes with Theoretical Detection Limits Naoya Ohmura,*,† Steve J. Lackie,‡ and Hiroshi Saiki†
Central Research Institute of Electric Power Industry, Department of Bioscience, 1646 Abiko, Abiko City, Chiba, Japan 270-11, Sapidyne Instruments, Inc., PMB 445, 967 East Park Center Boulevard, Boise, Idaho 83706-6700
A flow-based immunoassay that uses microspheres as the solid phase accomplished the theoretical limit of detectability achievable with the antibody. An equilibrated mixture of anti-estriol monoclonal antibody and estriol was briefly exposed to a bead pack containing immobilized estriol in a flow cell. A small portion of free antibody was separated rapidly from the mixture by binding it to immobilized hormone, but the antibody-hormone complex was kinetically excluded from binding. This rapid separation prevented shift in the equilibrium of the liquid phase binding. Signals were generated by labeling the separated antibodies on the beads with a Cy5-conjugated antispecies secondary antibody. By labeling after the separation step, perturbing the liquid-phase or solid-phase binding was prevented. This assay allowed the reduction of the concentration of primary antibody by continuously accumulating free antibody onto the beads prior to quantification and, thus, offered ideal conditions to achieve theoretical limits of detectability. The optimum achievable dynamic range of this immunoassay was 4-300 pM. Because the proportion of free anti-estriol antibody in the mixture was controlled by the Kd of the antibody-estriol interaction, when the concentration of the antibody was below the Kd, the smallest detectable estriol concentration approached the theoretical limit of detectability achievable with this antibody. Identification of environmental contaminants and their effects on various wildlife species has focused attention on the impact of environmental estrogens on human health.1 Xenoestrogens have created a demand for a highly sensitive assay capable of detecting very low levels of biologically active chemicals. Although the known immunoassays with the lowest detection limit are “excessreagent” assays,2 it is difficult to apply these assays to small analytes such as estrogenic chemicals, because they lack two antigenic sites that can be recognized simultaneously. The other category of immunoassay is “limited-reagent” assays, which employ single antigenic determinations.2 In contrast to excess* To whom correspondence should be addressed. Phone: +81-471-82-8211. Fax: +81-471-83-3347. E-mail:
[email protected]. † Central Research Institute of Electric Power Industry. ‡ Sapidyne Instruments Inc.. (1) Collborn, T.; vom Saal, F. S.; Soto, A. M. Environ. Health. Perspect. 1993, 101, 378-384. (2) Gosling, J. P. Clin. Chem. 1990, 36, 1408-1427.
3392 Analytical Chemistry, Vol. 73, No. 14, July 15, 2001
reagent assays, the ultimate sensitivity of limited-reagent assays is controlled by the affinity of the antibody.2,3 It is well-understood that at equilibrium, the law of mass action will largely dictate the relative proportions of bound and free antibody, which will further depend on the total amounts of antibody and antigen present.4 At higher levels of total antibody, the portion remaining free is governed by the total amount of antibody present. On the other hand, resolving the lowest possible levels of antigen requires minimizing the amount of antibody used in the assay. If, as a result, the total binding site concentration of antibody falls below the value of the dissociation constant (Kd), the fraction of antibody that is bound would be governed by the affinity of the antibody for its antigen and not by the amount of antibody present. Consequently, the portion of the antibody that would be existing free would be determined solely by the Kd and the concentration of antigen. In that regard, a practical detection limit must reach the theoretical limit when the Kd of antibody controls the binding in the assay. Although several developments of antibodies,5 immunoassay supports,6,7 related reagents,8 and assay configurations9 have been investigated for the past several years, there have been few studies to improve sensitivity by achieving conditions in which affinity controls the binding between a high-affinity antibody and antigen.10-12 Recently, an assay involving the use of a flow fluorometer with microspheres as the solid phase was reported to be a good method for measuring the binding constants of biologically active substances.13,14 One advantage of this method (3) Jackson, T. M.; Ekins, R. P. J. Immunol. Methods 1986, 87, 13-20. (4) Conners, K. A. Binding Constants: The Measurement of Molecular Complex Stability; Wiley: New York, 1987. (5) Hayden, M. S.; Gilliland, L. K.; Ledbetter, J. A. Curr. Opin. Immunol. 1997, 9, 201-212. (6) Piletsky, S. A.; Piletska, E. V.; Chen, B.; Karim, K.; Weston, D.; Barrett, G.; Lowe, P.; Turner, A. P. F. Anal. Chem. 2000, 72, 4381-4385. (7) Willumsen, B.; Christian, G. D.; Ruzicka, J. Anal. Chem. 1997, 69, 34823489. (8) Yazynina, E. V.; Zherdev, A. V.; Dzantiev, B. B.; Izumrudov, V. A.; Gee, J. S.; Hammock, B. D. Anal. Chem. 1999, 71, 3538-3543. (9) Hage, D. S. Anal. Chem. 1999, 71, 294R-304R. (10) Friguet, B.; Chaffotte, A. F.; Djavadi-Ohaniance, L.; Goldberg, M. E. J. Immunol. Methods 1985, 77, 305-319. (11) O’Connor, T.; Kane, M. M.; Gosling, J. P. Biochem. Soc. Trans. 1995, 23, 393S. (12) O’Connor, T.; Gosling, J. P. Biochem. Soc. Trans. 1998, 26, S42. (13) Blake, D. A.; Chakrabarti, P.; Khosraviani, M.; Hatcher, F. M.; Westhoff, C. M.; Goebel, P.; Wylie, D. E.; Blake, R. C. J. Biol. Chem. 1996, 271, 2767727685. (14) Blake, R. C.; Pavlov, A. R.; Blake, D. A. Anal. Biochem. 1999, 272, 123134. 10.1021/ac001328d CCC: $20.00
© 2001 American Chemical Society Published on Web 06/08/2001
is that the amount of antibody in the assay can be subpicomolar. In this report, we describe the optimized detection of female hormones using this assay under conditions in which affinity governed the amount of anti-estriol antibody bound to its antigen. The developed assay may be applied as a rapid and highly sensitive detection tool for quantifying other estrogenic chemicals existing in the environment. MATERIALS AND METHODS Materials. PMMA [poly(methyl methacrylate)] beads were from Sapidyne Instruments Inc. (Boise, ID); estriol and estriol6-(o-carboxymethyl)oxime:BSA were from Sigma (St. Louis, MO); DMSO (dimethyl sulfoxide) was from Wako (Tokyo, Japan); mouse anti-estriol monoclonal antibodies were from Biostride, Inc. (P01-92-31M-P; Redwood, CA) and Biodesign International (E45021M; Saco, ME); mouse anti-estradiol monoclonal antibody was from Biodesign International (H54069; Saco, ME); Cy5conjugated (fluorescent dye based on indodicarbocyanine) goat anti-mouse mAb was from Jackson ImmunoResearch (West Grove, PA). Preparation of Antigen-Coated Beads. PMMA beads with an average diameter of 100 µm were used in all of the experiments. Antigen was applied by immersing 200 mg of beads (dry wt) in 1.0 mL of PBS (pH 7.4) containing 100 µg of estriol-6-(ocarboxymethyl)oxime:BSA conjugate. After agitating for 2 h at 37 °C, the supernatant was removed. The beads were then immersed in 1.0 mL of PBS containing 10 mg/mL BSA and incubated at the same temperature for an additional 2 h to block nonspecific binding sites. The blocked beads were rinsed three times with 1.0 mL of PBS to remove any remaining reagents then suspended in 30 mL of PBS in the instrument’s bead reservoir. Assay. All experiments were performed with an automated flow immunoassay system that used beads as the solid phase (KinExA 3000, Sapidyne Instruments Inc.; Boise, ID). The system consisted of a capillary flow cell fitted with a microporous screen, which was integrated into an epi-illumination filter fluorometer system and through which flowed selected solutions under negative pressure created using a syringe pump (Figure 1A). The beads were packed into the flow cell by drawing 700 µL of bead suspension through the cell at a rate of 1.5 mL/min, washing it with 1.5 mL of PBS, and allowing it to settle for 15 s, thus creating a uniform bed. The immobilized estriol was then used to detect mouse anti-estriol mAb that was present in equilibrated mixtures of anti-estriol, estriol, and the antibody-antigen complex. To make the mixture, anti-estriol mAb was diluted to a concentration of between 20 and 2000 pM in PBS supplemented with 1.0 mg/mL of BSA to reduce nonspecific binding. Estriol was first dissolved to a concentration of 1 µM in DMSO; further dilutions were made in the BSA-supplemented PBS. Equal volumes of hormone and antibody solution were then combined, and once equilibrium was achieved (
