Anal. Chem. 2004, 76, 5393-5398
Capillary Electrophoresis Immunoassay Chemiluminescence Detection of Zeptomoles of Bone Morphogenic Protein-2 in Rat Vascular Smooth Muscle Cells Junhua Wang,† Weihua Huang,† Yanming Liu,‡ Jieke Cheng,*,† and Jing Yang§
Department of Chemistry, Wuhan University, Wuhan 430072, China, Department of Chemistry, Xinyang Normal University, Xinyang 464000, China, and Department of Pathology, Tongji Medical College of Huazhong University of Science and Technology, Wuhan, 430030, China
A capillary electrophoresis immunoassay (CEIA) method based on enhanced chemiluminescence (CL) detection was developed and applied to arteriosclerosis pathology research in the medical field. The system of enzymehorseradish peroxidase (HRP) catalyzing the luminol/ H2O2/p-iodophenol reaction was adopted in this paper. HRP was detected with the detection limit (S/N ) 3) of 4.4 pM (53 zmol), which represents one of the highest sensitivities of HRP reported yet. HRP was first linked to bone morphogenic protein-2 (BMP-2) in rat vascular smooth muscle (VSM) cells with noncompetitive format and analyzed by CE-CL. HRP-Ab2-mAb-BMP-2 complexes can be baseline separated from free HRP in 3 min. The detection limit (S/N ) 3) of BMP-2 is 6.2 pM (75 zmol). This technique has been successfully applied to arteriosclerosis development mechanistic study by investigating the change of BMP-2 content in VSM cells, which were stimulated by angiotensin II for different times. The change trends of BMP-2 contents are well in accord with that of the commonly used pathology image analysis system. It proves that the CEIA-CL technique proposed could be developed into a sensitive and new method for clinical assay and pathology research. Ultrasensitive analyses have potential applications in many research areas, such as chemical analysis, DNA sequencing, molecular dynamics, nanostructured materials, and early disease diagnoses.1 Immunoassays permit the highly sensitive and selective detection of macromolecular substances (e.g., proteins and polysaccharides) and of smaller molecules (e.g., peptides, hormones, and drugs) in complex biological matrixes, which cover an important field of analytical chemistry. Since Nielsen2 demonstrated the ability of capillary electrophoresis (CE) to separate an antigen-antibody complex from antibody and free antigen, CE * Corresponding author. Tel: +86-27-8768-2291. Fax: +86-27-8764-7617. E-mail:
[email protected]. † Wuhan University. ‡ Xinyang Normal University. § Tongji Medical College of Huazhong University of Science and Technology. (1) Lyon, W. A.; Nie, S. Anal. Chem. 1997, 69, 3400-3405. (2) Nielsen, R. G.; Rickard, E. C.; Santa, P. F.; Sharknas, D. A.; Sittampalam, G. S. J. J. Chromatogr. 1991, 539, 177-185. 10.1021/ac049891+ CCC: $27.50 Published on Web 08/19/2004
© 2004 American Chemical Society
has been proven to be a powerful separation tool and is therefore being examined as a useful separation method for rapid and efficient immunoassays in the past decade.3-5 The ultrasensitive capillary electrophoresis immunoassays are primarily based on laser-induced fluorescence (LIF) detection because of its broad applicability and high sensitivity. Measurements of human growth hormone,6 cortisol,7 insulin,8 and proteins9,10 have been reported. However, the high background due to the noise from Rayleigh and Raman scattering and the fluorescent impurities in the solvent are major disadvantages of LIF. An alternative ultrasensitive detection method is chemiluminescence (CL), which is characterized by simple and low-cost optical systems requiring no light sources, avoiding the effects of stray light and the instability of light source, and thus providing low background with excellent sensitivity. As an extremely useful analytical tool, CL gets an ever-widening number of novel uses in ultrasensitive analyses.11 Over the past decade, research in combing CE with CL detection has increased significantly.12-17 Great efforts have been initially made by improving CL systems18-24 (3) Schmalzing, D.; Buonocore, S.; Piggee, C. Electrophoresis 2000, 21, 39193930. (4) Heegaard, N. H. H.; Kennedy, R. T. J. Chromatogr., B 2002, 768, 93-103. (5) Yeung, W. S. B.; Luo, G. A.; Wang, Q. G.; Ou, J. P. J. Chromatogr., B 2003, 797, 217-228. (6) Shimura, K.; Karger, B. L. Anal. Chem. 1994, 66, 9-15. (7) Schmalzing, D.; Nashabeh, W.; Yao, X. W.; Mhatre, R.; Regnier, F. E. Anal. Chem. 1995, 67, 606-612. (8) Tao, L.; Aspinwall, C. A.; Kennedy, R. T. Electrophoresis 1998, 19, 403408. (9) Ou, J. P.; Wang, Q. G.; Cheung, T. M.; Chan, S. T. H.; Yeung, W. S. B. J. Chromatogr., B 1999, 727, 63-71. (10) Carnelley, T. J.; Barker, S.; Wang, H.; Tan, W.; Weinfeld, M.; Le, X. C. Chem. Res. Toxicol. 2001, 14, 1513-1522. (11) Roda, A.; Guardigli, M.; Michelini, E.; Mirasoli, M.; Pasini, P. Anal. Chem. 2003, 75, 463A-470A. (12) Garcı´a-Campan ˜a, A. M.; Baeyens, W. R. G.; Zhao, Y. Anal. Chem. 1997, 69, 83A-88A. (13) Baeyens, W. R. G.; Schulman, S. G.; Calokerinos, A. C.; Zhao, Y.; Garcı´aCampan ˜a, A. M.; Nakashima, K.; Keukeleire, D. De. J. Pharm. Biomed. Anal. 1998, 17, 941-953. (14) Garcı´a-Campan ˜a, A. M.; Gamiz-Gracia, L.; Baeyens, W. R. G.; Barrero, F. A. J. Chromatogr., B 2003, 793, 49-74. (15) Liu, Y. M.; Cheng, J. K. J. Chromatogr., A 2002, 959, 1-13. (16) Huang, X. J.; Fang, Z. L. Anal. Chim. Acta 2000, 414, 1-14. (17) Kuiper, C.; Milofsky, R. Trends Anal. Chem. 2001, 20, 232-240. (18) Dadoo, R.; Colo´n, L. A.; Zare, R. N. J. High Resolut. Chromatogr. 1992, 15, 133-135.
Analytical Chemistry, Vol. 76, No. 18, September 15, 2004 5393
and by improving the CE-CL interface design.18,20,22,25-28 In recent years, microchip-based CE-CL detection system has taken a page from the trees.29-32 Combination of CL detection with immunoassay is well-applied in bioanalytical and clinical chemistry,13,33-36 which can exert the ultrahigh sensitivity of CL and high selectivity of immunoassay. Kinds of diagnostic methodologies have evolved, with CL-based immunoassays becoming centrally important in the analysis of drugs, pesticides, hormones, and proteins.13 However, the study of CL immunoassays based on CE and microchip-CE is a first for us. Mangru and Harrison29 initially designed a novel microchipCE device for immunoassay combined with CL detection. Horseradish peroxidase (HRP)-fluorescein-labeled mouse IgG and free HRP-fluorescein were separated and detected with the detection limits for HRP of 9.0 nM. Similarly, Tsukagoshi et al.28 demonstrated a newly designed batch-type CL detection cell for CE. Experiments were performed to confirm the feasibility and superiority of the detection cell. They carried out a capillary electrophoresis immunoassay (CEIA) experiment using mouse IgG and HRP-labeled anti-mouse IgG antibody as a demonstration system and separated and detected HRP-labeled Ab and the complex. The detection limits for HRP are 2.0 nM. These previous works reveal that the CEIA-CL method may be a potential and promising analytical tool in biomedical analysis and clinical determination. Nevertheless, its feasibility and performance in actual applications should be further examined. In this atherosclerosis pathology-related study, we have demonstrated a noncompetitive immunoassay to determine the bone morphogenic protein-2 (BMP-2) by CE-CL using HRP-labeled Ab2-mAb and BMP-2 as the model. The abnormal proliferation of rat vascular smooth muscle (VSM) cells is a main step leading to atherosclerosis. It is adjusted by a number of growth factors and growth inhibitors. BMP-2, which belongs to the transforming growth factor superfamily, is a newly found growth inhibitor. In addition to bone induction, (19) Zhao, J. Y.; Labbe, J.; Dovichi, N. J. J. Microcolumn Sep. 1993, 5, 331-339. (20) Hara, T.; Okamura, S.; Kato, J.; Yokogi, J.; Nakajima, R. Anal. Sci. 1991, 7, 261-263. (21) Ruberto, M. A.; Grayeski, M. L. Anal. Chem. 1992, 64, 2758-2762. (22) Dadoo, R.; Seto, A. G.; Colo´n, L. A.; Zare, R. N. Anal. Chem. 1994, 66, 303-306. (23) Forbes, G. A.; Nieman, T. A.; Sweedler, J. V. Anal. Chim. Acta 1997, 347, 289-293. (24) Tsukagoshi, K.; Miyamoto, K.; Saiko, E.; Nakajima, R.; Hara, T.; Fujinaga, K. Anal. Sci. 1997, 13, 639-642. (25) Lee, Y. T.; Whang, C. W. J. Chromatogr., A 1997, 771, 379-384. (26) Huang, B.; Li, J. J.; Zhang, L.; Cheng, J. K. Anal. Chem. 1996, 68, 23662369. (27) Zhang, Y.; Gong, Z. L.; Zhang, H.; Cheng, J. K. Anal. Commun. 1998, 35, 293-295. (28) Tsukagoshi, K.; Nakamura, T.; Nakajima, R. Anal. Chem. 2002, 74, 41094116. (29) Mangru, S. D.; Harrison, D. J. Electrophoresis 1998, 19, 2301-2307. (30) Hashimoto, M.; Tsukagoshi, K.; Nakajima, R.; Kondo, K.; Arai, A. J. Chromatogr., A 2000, 867, 271-279. (31) Liu, B. F.; Ozaki, M.; Utmusi, Y.; Hattori, T.; Terabe, S. Anal. Chem. 2003, 75, 36-41. (32) Su, R. G.; Lin, J. M.; Qu, F.; Chen, Z. F.; Gao, Y. H.; Yamada, M. Anal. Chim. Acta 2004, 508, 11-15. (33) Thorpe, G. H. G.; Kricka, L. J.; Moseley, S. B.; Whitehead, T. P. Clin. Chem. 1985, 31, 1335-1341. (34) Roda, A.; Pasini, P.; Musiani, M.; Girotti, S.; Baraldini, M.; Carrea, G.; Suozzi, A. Anal. Chem. 1996, 68, 1073-1080. (35) Rongen, H. A. H.; Hoetelmans, R. M. W.; Bult, A.; Van Bennekom, W. P. J. Pharm. Biomed. Anal. 1994, 12, 433-462. (36) Dodeigne, C.; Thunus, L.; Lejeune, R. Talanta 2000, 51, 415-439.
5394
Analytical Chemistry, Vol. 76, No. 18, September 15, 2004
BMP-2 was found to be involved in the development of atherosclerosis.37 Further research demonstrated that BMP-2 has a potent inhibitory action on the growth of both rat and human VSM cells.38 On the other hand, angiotensin II(AgII), as the main effecter peptide of the rennin-angiotensin system, has played an important role in the pathogenesis of atherosclerosis.39 AgII can stimulate proliferation of VSM cells and led to inflammation through production of cellular growth factor. The effect of AgII on proliferation of VSM cells is an attractive subject in atherosclerosis pathology research. Its effect on a growth inhibitor such as BMP-2 is yet not clear. In the present work, we detected the BMP-2 content in VSM cells by CEIA with CL detection for the first time. The effects of pH, buffer concentration, and CL reagent concentration on sensitivity and resolution were also discussed. The changes of BMP-2 contents in VSM cells with AgII stimulation for different times were also investigated by CEIA-CL. EXPERIMENTAL SECTION Apparatus. The CE-CL apparatus was self-assembled in the laboratory. It is mainly based on the following parts: A highvoltage power source (0-30 kV, Peking University), a 3066-type chart recorder (The Fourth Instrumental Factory of Sichuan, China), a photomultiplier tube (R928, Hamamatsu Photonics), and a type HX-2 signal magnifier (Institute of Chemistry, Chinese Academy of Sciences, Beijing, China). Three kinds of fused-silica capillaries (Hebei Optical Fiber Factory) were used in the device: a separation capillary (50-µm i.d., 375-µm o.d.), a reaction capillary (530-µm i.d.), and an introducing capillary (320-µm i.d.). The scheme of the CE-CL apparatus is shown in Figure 1, where the description of the device can be found. We used an inverse microscope (Axiovert 200M, Zeiss) under 40×, 0.6 NA to observe the cells and a high resolving power digital camera to image (3900 × 3090 pixel, AxioCam, Zeiss). A type HPIAS 2000 Pathology Image Analyse System was used to perform colorimetric detection. Cell scraper, sonic washer, and centrifuge were used in making sample solutions from cell slides. Reagents and Materials. Luminol was obtained from the Chemistry Department of Shanxi Normal University (Shanxi, China), p-iodophenol (PIP), 30% hydrogen peroxide solution, from Shanghai Chemical Factory (Shanghai, China), and HRP from the Sino-American Biotechnology Co. Ltd. (Wuhan, China). All chemicals were of analytical-reagent grade. The 18.2 MΩ/cm water was purified with a Water PRO PS system (Labconco, Kansas City, KS). VSM cells (purity >90%) were harvested from 4-6-week-old male Sprague-Dawley rats, which were provided by Tongji Experimental Animal Center (Wuhan, China). The cells contain BMP-2 were stimulated by AgII (Sigma, St. Louis, MO) for 6, 12, and 24 h and harvested on glass slides. Rabbit anti-rat BMP-2 monoclonal antibody (mAb) was obtained from Borston Co. (Wuhan, China). Rabbit streptavidin/peroxidase kit (including biotin-goat anti-rabbit-IgG, blocking solution, and streptavidin (37) Willette, R. N.; Gu, J. L.; Lysko, P. G.; Anderson, K. M.; Minehart, H.; Yue, T. L. J. Vasc. Res. 1999, 36, 120-125. (38) Nakaoka, T.; Gonda, K.; Ogita, T.; Otawara-Hamamoto, Y.; Okabe, F.; Kira, Y.; Harii, K.; Miyazono, K.; Takuwa, Y.; Fujita, T. J. Clin. Invest. 1997, 100, 2824-2832. (39) Weiss, D.; Sorescu, D.; Taylor, W. R. Am. J. Cardiol. 2001, 87, 25C-32C.
Figure 2. Optical micrographs of VSM Cells. (A) Before adding mAb. (B) After adding Ab2 and HRP to form BMP-2-mAb-Ab2*. Observed under 40×, 0.6 NA with Axiovert 200 M Zeiss inverse microscope.
Immune Reaction Procedure. The immunoreactions were conducted as follows which are similar to that adopted by Bernard.40
BMP-2 + mAb (excess) f BMP-2-mAb + mAb (1) Figure 1. Scheme of the CE-CL apparatus self-assembled in our laboratory, which belongs to off-column coaxial flow mode. Reaction tee reservoir was made in plexiglass with a baffle in the middle. The CL reagents were introduced through the hole in the baffle. The cathode was placed in the same hole. The PMT was fluted in a slot in the inside wall to allow the reaction capillary go across its light collection window. A 1-cm detection window was made on a suitable place of reaction capillary. The detection window was placed just in front of the PMT’s light collection window. The outlet of separation capillary was also placed before the window where the analyte and the CL reagent take reactions. The whole device was held in a lighttight box to exclude stray light.
HRP) was purchased from Zhongshan Biology Engineering Ltd. (Beijing, China). Preparation of 1.0 × 10-2 M luminol stock solution (prepared every two weeks): 0.1772 g of luminol was dissolved in 0.1 M NaOH, then diluted with water in a 100-mL brown flask, and stored in the refrigerator. Preparation of 1.0 × 10-4 M HRP stock solution (prepared every week): 0.0044 g of HRP dissolved with water in a small vial and refrigerated for use; the lower concentration solutions used every day were diluted from it. Preparation of 1.0 × 10-2 M PIP stock solution (prepared every week): 0.2200 g of PIP was dissolved with water in a 100-mL flask, a small amount of ethanol was added to dissolve it, and then water was added to volume. The electrophoretic buffer, NaH2PO4-Na2HPO4 buffer (pH 6.5) was filtered through a 0.22-µm pore-size membrane prior to use. The postcolumn CL reagent was the mixture of luminol, H2O2, and PIP (pH 10.5), prepared to the needed concentration before use, and pH adjusted by 1.0 M NaOH and HNO3. Before the experiment, all of the vessels and tips for pipetting were dipped in 3.6 M HNO3 for 48 h and then washed with ultrapure water. Treatment of Capillary. New separation capillary was treated with 2.0 M NaOH-CH3OH, 2.0 M NaOH, 1.0 M HCl, and water in sequence prior to use and balanced with electrophoretic buffer overnight. During every run, the capillary was washed with buffer for 3-5 min. Separation capillary, reagent guiding capillary, reaction capillary, and tee cell were all injected with individual solution by syringe.
BMP-2-mAb + Ab2 (excess) f BMP-2-mAb-Ab2 + Ab2 (2) BMP-2-mAb-Ab2 + HRP (excess) f BMP-2-mAb-Ab2* + HRP (3) The slides were placed horizontally, 1% TritonX-100 was added, and the slides stayed at room temperature for 15 min. The cells’ velum was broken. The slides were washed with 0.01 M phosphate buffer saline (PBS, pH 7.4) three times. Goat serum containing 3% H2O2 was added to block the nonspecific site of rabbit anti-rat BMP-2-mAb and to inactivate the endogenous peroxidase activity, and the cells were washed with PBS for 3 times. First antibody (rabbit anti-rat mAb) was added to cover the whole slide, and the slide was incubated at 4 °C overnight; mAb was linked to BMP-2 as shown in eq 1. The slide was washed with PBS three times. Then secondary antibody was added; the slide was incubated at 37 °C for 2 h and then washed with PBS for an additional three times. Ab2 specifically reacted with mAb and was linked to the complex as shown in eq 2. Then 5 µL of 1.0 × 10-4 M HRP was added and the resultant mixture was incubated at 37 °C for 2 h. HRP was marked on the biotin-secondary antibody to form BMP2-mAb-Ab2-HRP (BMP-2-mAb-Ab2*) as in eq 3. Cells Slide Selection and Sample Preparation. Before the immunoassay experiments, we selected the cells’ slides carefully under the microscope. The slides with proportional distributing density (see Figure 2) and cells amounts of 400 ((5) on whole slide were used. After labeling, we examined the integrality of cell shape and amount and selected the cells in good shape and a strict amount of 400 ((5). The process of sample preparation was conducted as described in the literature.41 The sample carefully scraped with a cell scraper to remove the adherent cells from the slides. Then they were dissolved in 50 µL of water. The solution was sonicated for 15 (40) Bernard, A.; Michel, B.; Delamarche, E. Anal. Chem. 2001, 73, 8-12. (41) Simpson, R. J. Proteins and Protomics: A Laboratory Manual; Cold Harbor Laboratory Press: Cold Harbor, NY, 2003; pp 165-166.
Analytical Chemistry, Vol. 76, No. 18, September 15, 2004
5395
Figure 3. Effect of pH (A), concentration of H2O2 (B), luminol (C), and PIP (D) on the CL intensity. The concentration of HRP used for optimal conditions was 1.0 × 10-8 M. Average values are plotted, and one standard deviation (1σ) is indicated by error bars (n ) 5).
min and centrifuged at 5000g for 5 min; the supernatant was divided into 50-µL vials as the analytical samples. To guarantee that the 50-µL sample solutions contain the same amount of cells, it is important and necessary to observe and select the cell slides. Figure 2 shows that we have selected the slides containing little difference in amount and in shape before and after the immunoreactions. RESULTS AND DISCUSSION Effects of pH and Concentration of Running Buffer. The effect of pH and concentration of running buffer was investigated in detail. The results obtained show that pH 6.5 is the best pH value at which the peaks are sharp and symmetrical. The possible reason is that, as a small molecular glycoprotein, the isoelectric point of HRP is 6.5. When it is below pH 6.5, the positive-charged HRP would be absorbed to the capillary wall. When the pH is over 6.5, it would be too far from the optimal condition of HRP catalysis activity (pH 3-4). Comparing the PBS concentrations of 0.01, 0.02, and 0.05 M, we found that the electrophoretic current was larger than we expected when 0.02 and 0.05 M buffers were used. The reason may be that the Joule heating becomes larger with the increase of electrophoretic buffer concentration, which results in a larger electrophoretic current in the capillary. Optimization of the Conditions for Postcolumn CL Solution. We studied in detail the influence of pH and concentration of H2O2, luminol, and PIP using the HRP concentration of 1.0 × 10-8 M for the experimental univariate. The pH value of the CL reagent has a great influence on CL intensity. We studied the pH from 7 to 12, Figure 3A show that when the pH is over 11.0, background noises greatly increase and 5396 Analytical Chemistry, Vol. 76, No. 18, September 15, 2004
the signal-to-noise ratio drops. Therefore, pH 10.5 was chosen as the optimal pH value in our experiments. We also examined the optimal H2O2 concentration. The signal increases with the increase in H2O2 concentration. Through Figure 3B, we can see the increase gets slower and S/N declines dramatically as it reaches 30 mM. So 30 mM was chosen as the reasonable concentration. We examined the effect of luminol concentration from 1.0 × 10-5 to 1.0 × 10-3 M and found that when it was at 5.0 × 10-4 M, the curve of intensity became flat (Figure 3C). We used this concentration in the experiments. In the reaction of luminol-H2O2 catalyzed by HRP, PIP is an effective enhancer in para-substituted phenol compounds.33 We studied the PIP concentration from 0 to 2.5 mM and found that when it was 1.25 mM, it exhibited the best enhancing effect. The signal got weaker when the concentration increased (Figure 3D). Reasonable explanation for this phenomenon is unclear so far. Untrasensitive Detection of HRP. Using PIP as an enhancer for the luminol-H2O2 system, the reaction has been applied to the detection of HRP. Under optimized conditions, the ultrasensitive detection of HRP was realized. Figure 4A shows electropherograms of two injections of 1.25 × 10-11 M HRP. According to the peak height, the limit of detection is 4.4 × 10-12 M (S/N ) 3), while the mass detection limit is 5.3 × 10-20 mol (53 zmol) in terms of 12-nL (10 kV, 10 s) injection volume. According to experimental data, the intraday relative standard deviation (RSD) of migration time is 1.2% (n ) 5). The RSD of peak height is no more than 5.5% (n ) 5). The linear range of HRP is 5.0 × 10-111.0 × 10-8 M (correlation coefficient, 0.998). This sensitivity is
Figure 4. Electropherograms of (1) HRP and (2) Ab2*-mAb-BMP-2 in VSM cells. (A) 1.25 × 10-11 M HRP standard two-time injection. (B) BMP-2 in VSM cells without AgII stimulation. (C) AgII stimulating VSM cells for 6 h. (D) AgII stimulating VSM cells for 12 h. (E) AgII stimulating VSM cells for 24 h. Capillary, 55-cm length and 50-µm i.d. of fused silica, applied running voltage, 20 kV; injected voltage 10 kV, 10 s; migration buffer, phosphate buffer (10 mM, pH 6.5). In the postcolumn cell, 30 mM H2O2, 0.5 mM luminol, and 1.25 mM PIP, pH 10.5.
among the highest detection sensitivities for HRP using CL detection. 11 Detection of BMP-2 in VSM Cells. Determination of BMP-2 in rat or human VSM cells has created great interest in the atherosclerosis research field. By conventional enzyme-linked immunosorbent assay (ELISA) method, semiquantitative analysis of the bulk or mean of thousands of cells could be achieved. In this experiment, we analyzed a small amount of cells with CE in order to test the feasibility, sensitivity, and quantifiability of the CEIA-CL method. The normal cells were made into sample solutions and injected into the capillary. Figure 4B shows the electropherograms of the sample solutions containing HRP and BMP-2-mAb-Ab2*, which showed that free HRP and the complexes were baseline separated in 3 min by CE. According to the migration time of the HRP standard (Figure 4A), we could easily discriminate the peaks of HRP and the complexes in electropherograms. BMP-2 is detected with the limit of 6.2 pM (75 zmol). Application in Pathology Research. We have carried out this cooperative study on atherosclerosis pathology with Tongji Medical College to examine how the BMP-2 content in VSM cells changes when AgII stimulates cells for different times. To compare the results with routine ELISA tests, the materials we used, experimental strategy design, and immunoassay steps in our experiments were partly the same as that in the routine immunohistochemistry methods.
Table 1. Relative CL Intensity of HRP-Ab2-mAb-BMP-2, with AgII Stimulation for Different Timesa time (h) 6
12
group
intensb
RSD (%)
normal comparison Ag stimulating
121 138
3.8 4.6
a
24
intensb
RSD (%)
intensb
RSD (%)
122 64
4.1 4.4
123 23
3.9 4.5
The RSD was calculated for n ) 5. b Relative CL intensity.
Based on the ultrasensitive detection of BMP-2, we analyzed the sample solution of cells stimulated by AgII for 0, 6, 12, and 24 h. From the electropherograms (B-E in Figure 4), we can see that the peak heights of BMP-2-mAb-Ab2* changed obviously with stimulation time. It indicates that AgII could have great influence on the content of BMP-2. According to the previous detection, BMP-2 in AgII-treated cell samples were reckoned within the 10.0-1.0 pM range. We also performed parallel immunoassays that replaced the scraping step with staining by 3′,3-diaminobenzdine tetrahydrochloride solution, followed by microscopic observation. The finial photometric results were detected with the HPIAS 2000 Pathology Image Analyse System Analytical Chemistry, Vol. 76, No. 18, September 15, 2004
5397
should be taken for consideration in atherosclerosis pathology research.
Figure 5. Curve of relative BMP-2 expression in VSM cells with AgII stimulation, for 0, 6, 12, and 24 h, detected by CE-CL (A) and by HPIAS 2000 (B). Average values are plotted, and one standard deviation (1σ) is indicated by error bars (n ) 5).
Figure 4B-E, Table 1, and curve A in Figure 5 were the detection results based on our CEIA-CL method. Curve B in Figure 5 were that of immunohistochemistry method. Though determined by two different methods, the results are well in accord with each other. The results indicate that the content of BMP-2 increases with the AgII stimulation for 6 h, whereas a longer stimulation, such as 12 and 24 h or more, makes it decline continuously. The possible reason may be that AgII can only boost BMP-2 content for a short period (6 h). After which, it would act contrarily on BMP-2. The detailed explanation
5398 Analytical Chemistry, Vol. 76, No. 18, September 15, 2004
CONCLUSION We have developed a CL-based CEIA system with high sensitivity and resolution, which offers a novel technique in investigating the stimulating effect of AgII on the BMP-2 content in cells and the possible molecular mechanism for pathology research. The sensitivity and detectability of CL immunoassays are similar to or better than that of radioisotopes.11 In addition, CE-based CL immunoassays would be more effective and quantitative in studying complex biological microenvironments than traditional ELISA. This method would greatly enlarge the area and target of pathologic research. In a further study, we will focus our attention on the miniaturization, automation and on-line detection, and more sensitive device of CE-based immunoassays with CL detection. A chip-based CEIA-CL strategy should be more attractive. Its realization would accelerate CE technique in the field of clinical analytical chemistry as well as pathology research. ACKNOWLEDGMENT We thank Miss Ren Xinyu in Tongji Medical College greatly for providing the materials and performing parts of the immunoreactions, and we thank Miss Zhao Shan for the drawing of apparatus scheme in 3ds max soft. This work was supported by the National Natural Science Foundation of China (20299034). Received for review January 17, 2004. Accepted July 7, 2004. AC049891+
