Quantitative analysis of serum procollagen type I C-terminal propeptide by immunoassay on microchip

Abstract

Background: Sandwich enzyme-linked immunosorbent assay (ELISA) is one of the most frequently employed assays for clinical diagnosis, since this enables the investigator to identify specific protein biomarkers. However, the conventional assay using a 96-well microtitration plate is time- and sample-consuming, and therefore is not suitable for rapid diagnosis. To overcome these drawbacks, we performed a sandwich ELISA on a microchip. Methods and Findings: The microchip was made of cyclic olefin copolymer with straight microchannels that were 300 μm wide and 100 μm deep. For the construction of a sandwich ELISA for procollagen type I C-peptide (PICP), a biomarker for bone formation, we used a piezoelectric inkjet printing system for the deposition and fixation of the 1st anti-PICP antibody on the surface of the microchannel. After the infusion of the mixture of 2.0 μl of peroxidase-labeled 2nd anti-PICP antibody and 0.4 μl of sample to the microchannel and a 30-min incubation, the substrate for peroxidase was infused into the microchannel; and the luminescence intensity of each spot of 1st antibody was measured by CCD camera. A linear relationship was observed between PICP concentration and luminescence intensity over the range of 0 to 600 ng/ml (r2 = 0.991), and the detection limit was 4.7 ng/ml. Blood PICP concentrations of 6 subjects estimated from microchip were compared with results obtained by the conventional method. Good correlation was observed between methods according to simple linear regression analysis (R2 = 0.9914). The within-day and between-days reproducibilities were 3.2-7.4 and 4.4-6.8%, respectively. This assay reduced the time for the antigen-antibody reaction to 1/6, and the consumption of samples and reagents to 1/50 compared with the conventional method. Conclusion: This assay enabled us to determine serum PICP with accuracy, high sensitivity, time saving ability, and low consumption of sample and reagents, and thus will be applicable to clinic diagnosis. © 2011 Yatsushiro et al.