A Capillary Gel Electrophoresis Fluorescence System for Antibody–Antigen Interactions

Since its inception more than 20 years ago,1 slab gel-based electrophoresis technology has been used routinely for the analysis of biomolecules (i.e., DNA, proteins, and carbohydrates). However, use of slab gel electrophoresis in bioseparations applications is labor intensive and in need of improvement in terms of resolving power, throughput, and cost per sample.

Capillary gel electrophoresis (CGE) is a microfluidic approach and as such incorporates a microchannel device that simplifies gel electrophoresis. As an instrumental approach to gel electrophoresis, capillary electrophoresis (CE) offers a diverse range of applications. CE technology is widely accepted by the biotechnology industry, specifically in nucleic acid-based testing, as a reliable, high-resolution, high-sensitivity detection tool.2–4 Applications of capillary electrophoresis include oligonucleotide analysis, DNA sequencing, and dsDNA fragment analysis.2–6

In routine analysis, CE is often avoided because it is reputed to be troublesome, with high failure rates. However, this is no longer true now that instrument manufacturers have improved instrument design and overall CE knowledge has increased. Three key contributors reduce failure rate and allow accurate, precise, and robust CE data: operator training, system stability, and ease of instrument operation with low maintenance.7

Capillary electrophoresis immunoassay analysis

Capillary electrophoresis immunoassay analysis (CEIA) is a new analytical technique which, when combined with sensitive detection methods such as laser-induced fluorescence (LIF), offers several advantages over conventional immunoassays.8 CEIA can perform rapid separations with high mass sensitivity, simultaneously determine multiple analytes, and is compatible with automation. CE and fluorescent-labeled peptides can be used to detect abnormal prion protein in the blood of animals. One such CE-based noncompetitive immunoassay for prion protein using fluorescein isothiocyanate (FITC)-labeled Protein A as the fluorescent probe method was successfully applied for testing blood samples from scrapie-infected sheep.9,10

Immunoassays are commonly used in biotechnology for the detection and quantification of host cell contaminants. The free-solution approach by CE with fluorescence detection offers an exciting alternative to solid-phase immunoassay. It eliminates antigen immobilization and avoids many problems associated with solid phase. The methodology makes use of either a purified antigen labeled with stable fluorescent dye (i.e., FITC) or an affinity probe labeled with the dye (direct assay).11–13

CE and laser-induced fluorescence

CE with LIF is one of the most powerful analytical tools for rapid, high-sensitivity, high-resolution dsDNA analysis14,15 and immunoassay analysis applications. However, CE-based LIF systems are much more expensive than traditional slab gel-based bioanalysis systems. Thus they are out of the reach of all but a few well-funded laboratories and present a high-cost barrier to the expansion of immunoassay analysis applications.

To address the need for a system that is easy to use, economical, and provides rapid analysis with high efficiency, sensitivity, throughput, and standardization for CE-based applications, a single-channel CE instrument was developed for fluorescent immunoassay separation and detection that can be used in both research and clinical diagnostics laboratories.

Affinity capillary electrophoresis (ACE) was applied to analyze antigen–antibody interactions by using polyclonal antibodies against crocalbin, a calcium-binding protein with EF-hand motif, as a model system. Basically, there are two parameters to determine whether the immunoreaction forms the immunocomplex: 1) peak area or peak height, and 2) pattern or migration time of the peak. Using FITC-labeled protein sample, a sensitivity limit of ~15 nM or less was attained for the capillary electrophoresis immunoassay platform.

Materials and methods

CE instrumentation

The I-CE instrument design (Figure 1) (BiOptic, Inc., La Crescenta, CA) is based on real-time fluorescence detection. It features an economical, disposable, pen-shaped gel cartridge (Figure 2) for microseparations and fluorescent-labeled antibody fragment detection. The instrument’s modular design offers flexibility in single-channel CGE applications. It provides significant background noise reduction, resulting in improved S/N for high detection sensitivity for immunoassays at a low cost per sample.

 Figure 1 – I-CE instrument.
 Figure 2 – Components of single-channel gel cartridge.

The pen-shaped disposable gel cartridge incorporates a single microfluidic glass capillary in an injection-molded body with integrated gel reservoir (Figure 3). The unit includes top and bottom electrodes (anode and cathode), an exposed detection zone (window), and an embedded radiofrequency identification (RFID) label to track the number of runs and provide identification of the gel cartridge.

Figure 3 – Close-up of gel cartridge being installed in the instrument.

 Once the gel cartridge has been installed inside the instrument, an N2 or air-operated fork assembly (pneumatic actuated) aligns two optical fibers precisely to the detection zone of the glass capillary (Figure 4). Multimode optical fibers (100–500 μm) deliver the excitation light (from royal blue LED = 466 nm) and collect the emission signal (fluorescence light = 520 nm) and transfer it to a photomultiplier tube (PMT) for data analysis. The PMT module has a built-in emission filter (bandpass filter) to improve detection sensitivity.

Figure 4 – Gel cartridge engaged with fork assembly for detection.

The fully automated I-CE instrument includes an in-house-designed, automated, modular X-Y-Z mechanism (with axes for horizontal, vertical, and rotational motion) for the buffer and sample tray, which accepts either a 12-well sample strip or 96-well microtiter plate interfaced with the disposable single-channel gel cartridge. The biomolecules are injected into the gel cartridge by means of electrokinetic (high-voltage) injection. As the lower electrode (cathode) with its embedded capillary tubing is lowered inside the sample well, the applied high voltage electrokinetically injects the negatively charged biomolecules inside the capillary tubing; the biomolecules are then separated by the separation buffer migrating through the glass tubing toward the gel reservoir, where the anode (positive electrode) is located. The fluorescently tagged biomolecules are then detected along the length of the capillary at the detection zone by the emission collection optics. The modular fork assembly (Figure 4) provides accurate optical alignment and increases detection sensitivity, with reproducible results for fluorescence CE applications.

Chemicals and buffers

The following analytical-grade chemicals (Sigma, St. Louis, MO) were used: fluorescein isothiocyanate isomer 1 (FITC), 3-morpholinopropane-1-sulfonic acid (MOPS), tris (hydroxymethyl) aminomethane (tris), glycine, acrylamide, and hexadecyltrimethylammonium bromide (CTAB). Gel filtration with Sephadex® G-25 (Pharmacia Biotech, Piscataway, NJ) or dialysis using Spectra/Por porous membrane (Spectrum Laboratories, Rancho Dominguez, CA) was applied to eliminate the free FITC. Crocalbin, as antigen, was cloned in expression vector, and the recombinant protein in E. coli was purified according to the manufacturer’s protocol (QIAGEN, Germantown, MD). The hemocyanin-conjugated peptide, corresponding to one of the epitopes of crocalbin, was used to produce antibodies from rabbit. The IgGs, also known as poly(A)-binding protein (pAb), were isolated from the antiserum via Protein A affinity resin. Poly(ethyleneoxide) (PEO) sieving medium was purchased from a local chemical supplier.

Protein conjugated with FITC

The manufacturer’s instructions were followed with some modifications. Briefly, protein in 0.1 M sodium carbonate–bicarbonate buffer, pH 9.5, and FITC powder in dimethylsulfoxide (DMSO) were continuously mixed for 3 hr at room temperature under dim light. The reaction was stopped by adding 0.1 M tris buffer, pH 8.0, containing 1 M glycine. The DMSO concentration in the reaction was 5%, and the concentration of protein was ~2–4 mg/mL. To eliminate free FITC, the reaction mixture was either dialyzed or gel-filtered in phosphate-buffered saline. Both the protein concentration and FITC/protein molar ratio of the sample were measured at absorbances of 280 and 495 nm, respectively.

Free-solution capillary electrophoresis (FSCE)

All experiments were performed at room temperature (~25 °C). The detector on the anodic side was equipped with an LED light source with excitation at 488 nm and emission at 520 nm. The uncoated fused-silica capillary used was 16 cm in length (11 cm effective detection length) × 75 μm i.d. To perform FSCE, 50 mM glycine buffer, pH 9.0, containing 0.5 mM CTAB was used in both electrodes.16 On the cathodic side, the sample was electrokinetically injected and separation was conducted at 11 kV. Each experiment was performed at least twice. To determine the optimal reaction time for the immunocomplex formation, pAb (6.84 μM) and FITC-crocalbin (12.25 μM) were combined in the electrophoretic buffer with a final volume of 20 μL and incubated at room temperature. At the indicated times (0–8 hr), the reaction mixture was assayed. To determine the dose-dependent effect of pAb on the immunocomplex formation, the fixed amount of FITC-crocalbin (12.25 μM) and the increased amount of pAb (0.17–6.84 μM) were incubated together in the buffer. To ensure that the reaction reached equilibrium, it was placed at 4 °C overnight and then analyzed in the I-CE.

Results and discussion

To demonstrate the immunocomplex formation between pAb and crocalbin, both time-course and dose-dependent experiments were conducted. In the time-course experiment (Figure 5), during the first 4 hr the reaction mixture in the assay showed only one symmetrical sharp peak; after 4 hr, a new peak appeared. Under CE conditions, the new peak first appeared at 4.5 hr and then gradually reached equilibrium. Because the sampling mechanism is electrokinetic, the relationship among these patterns is not representative of a qualitative versus quantitative method. Before identification, it was assumed that the new peaks were the immunocomplex of pAb and crocalbin. The pAb dose-dependent effect on formation of the immunocomplex also showed that, as the concentration of pAb increased, the new peak increased in area and height. However, the maximum ratio of pAb to crocalbin was approximately 1:2 and the peak areas of free crocalbin and immunocomplex were almost equally distributed (Figure 6).

 Figure 5 – Time-course experiment for immunocomplex reaction in FSCE. Electropherograms present the migration-shift pattern of the immunocomplex formation. The molar ratio of FITC-crocalbin to pAb is ~1.8– 1.0. Electrophoretic conditions are described in text.
Figure 6 – Dose-dependent effect of antibody on immunocomplex formation in FSCE. Electropherograms present the separation of free FITC-crocalbin from pAb-bound FITC-crocalbin by ACE. A fixed amount of FITC-crocalbin (12.25 μM) and various amounts of pAb were used as indicated in the figure. A control experiment was performed in the absence of pAb.

In order to obtain more sensitive signals, glycine buffer at pH 9.0 was used. At this pH, FITC presents the highest relative fluorescence intensity. In addition, cationic detergent—CTAB—was used in the experiments to partially inhibit the electroosmotic flow (EOF) and reduce protein adsorption.17,18 However, CTAB seems to play dual, opposite roles: One inhibits EOF by neutralizing the negative charges located on the inner face of the capillary, and the other shows its capability for binding protein samples, causing it to be less negatively charged. CTAB may be considered a nondenaturing detergent because it retains the native activity of protein samples.

Figure 7 – Detection limit test for the protein sample in CGE. Electropherograms indicate the sensitivity of the I-CE by detecting the serial dilutions of FITC-BSA. The assays were analyzed in 25 mM MOPS-tris buff er, pH 8.9, containing 0.8% PEO (MW 7 × 106) or 2.5% polyacrylamide.

Protein quantitation by the I-CE was carried out using the measurement of FITC-BSA (bovine serum albumin). Either linear polyacrylamide or PEO was used as the matrix. The detection range increased to 333 nM with a detection limit of 15.6 nM (Figure 7).


The compact (15” L × 12” W × 16” H) I-CE design platform is able to rapidly resolve the reaction mixtures of antibody and antigen in a free-solution capillary electrophoresis system. A CGE system was used to separate Ag-pAb from Ag (data not shown). The detection range was between submicromolar and nanomolar; the lower detection limit was below the nanomolar level. The I-CE instrument can perform both noncompetitive immunoreaction assays and protein quantitation under native conditions. Future applications will include monitoring the conformational change of protein during the refolding process, and studying the mechanism of the multiple molecular complex responsible for specific biological phenomena.


  1. Maniatis, T.; Fritsch, E.F. Molecular Cloning: A Laboratory Manual; Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY, 1983.
  2. Guttman, A.; Cook, N. Anal. Chem.  1991, 63, 2038–42.
  3. Huang, X.C.; Quesada, M.A. et al. Anal. Chem.  1992, 64, 2149–54.
  4. Strege, M.; Lagu, A. Anal. Chem.  1991, 63, 1233–6.
  5. Amirkhanian, V.D.; Liu, M-S. Biomedical Nanotechnology Architectures and Applications, Proceedings of SPIE  2002, 4626, 238–45.
  6. Liu, M-S.; Amirkhanian, V.D. Electrophoresis  2003, 24, 93–5.
  7. Solano, S.O.; Gennaro, L. et al. CE currents. LC-GC Eur.  2008, 615–22.
  8. Bao, J. J.Chromatogr. B  1997, 699, 463–80.
  9. Yang, W.-C.; Schmerr, J.M. et al. Anal. Chem.  2005, 77, 4489–94.
  10. Schmerr, M.J.; Jenny, A. J. Chromatogr. 1999, 853, 207–14.
  11. PACE/Setter, CE in Biotechnology. The News letter for Capillary Electrophoresis; Spring  1998, 2(2).
  12. Schmerr, M.J.; Goodwin, K.R. et al. J. Chromatogr. B  1996, 681, 29–35.
  13. Shimura, K.; Karger, B.L. Anal. Chem. 1994, 66, 9–15.
  14. Schwartz, H.E.; Ulfelder, K. J. Anal. Chem.  1992, 64, 1737–40.
  15. Clark, S. M.; Mathies, R.A. Anal. Biochem.  1993, 215, 163–70.
  16. Chiu, T.-C.; Tu, W.-C. et al. Electrophoresis  2008, 29, 433–40.
  17. Akins, R.E.; Levin, P. et al. Anal. Biochem1992, 202, 172–8.
  18. Verzola, B.; Sebastiano, R. et al. Electrophoresis  2003, 24, 121–9.

Shou-Kuan Tsai is President, and Ming-Jhy Hseu is Technical Advisor, BiOptic, Inc., Taiwan County, Taiwan, R.O.C. Varoujan D. Amirkhanian is Technical Advisor, BiOptic, Inc., 3831 El Caminito St., La Crescenta, CA 91214, U.S.A.; tel.: 818-679-4413; fax: 818-679-4413; e-mail: varoujamirkhanian@bioptic.com.tw.