Post-column reaction detection based on fluorescence energy transfer in the far red spectral region

Coupling immunoassays with chromatographic separation techniques

Coupling immunoassays with HPLC separation techniques is becoming increasingly useful in the analysis of biological and nonbiological samples of both large and small molecules. This is because it provides both sensitivity and selectivity for molecular analysis at relatively low cost, low maintenance and with excellent potential for automation. This paper reviews application of this hyphenated approach both in the pre-column immunoextraction and post-column immunodetection modes. Systems in which immunoassays are interfaced to chromatographic separations in order to separate bound and free fractions of the immunoassay will not be included since these systems do not provide the enhanced selectivity common to hyphenated systems. Post-column immunodetection is based on various immunoassay formats such as direct detection, one-site, competitive and sandwich immunoassays. Homogeneous immunodetectors are more convenient than heterogeneous immunodectors since there are no separation and column regeneration steps involved in homogeneous immunoassays. On the other hand, heterogeneous immunoassays are generally more sensitive than homogeneous immunoassays since interfering substances are removed prior to immunodetection. Advantages and limitations for the various approaches will be discussed.

High-performance liquid chromatography with on-line post-column immunoreaction detection of digoxin and its metabolites based on fluorescence energy transfer in the far-red spectral region

The combination of immunoassays with separation techniques such as chromatography can result in enhanced selectivity and sensitivity. This paper describes an on-line chromatography with immunochemical post-column fluorescence energy transfer detection for digoxin and its metabolites. R-phycoerythrin (PE) was used as the donor and an indodicarbocyanine dye (Cy5) as the acceptor label. These labels allow the detection in the far-red spectral region, which is more selective for biological samples. Hence, digoxin was labeled with PE using the activated digoxigenin-NHS-ester and monoclonal anti-digoxin antibody was labeled with Cy5. Digoxin and its metabolites was injected into the HPLC system followed by post-column injection of R-phycoerythrin labeled digoxin and by Cy5 labeled anti-digoxin antibody. Incubation time was provided using an open tubular reactor coil at room temperature. The detection was performed by measurement of the sensitized emission of Cy5 at 670 nm due to fluorescence energy transfer from PE labeled with digoxin. The system was optimized with regard to the concentrations of the used post-column reagents as well as incubation time and temperature. The dynamic range of digoxin spiked in 0.01 M phosphate buffer (pH 7.4) was 0.05 to 10 ng/ml with a correlation coefficient of 0.989. The limit of detection was 33 pg/ml. The precision of two controls, 0.4 and 4 ng/ml, was found to be 2.2 and 8.7% RSD, respectively, accuracy was 10.7 and 20.3% (n=6 in each case).

Post-column reaction detection of biotin in human plasma ultrafiltrate based on laser-induced fluorescence energy transfer in the far-red spectral region

High performance liquid chromatography followed by post-column reaction detection in the far-red spectral region provides added sensitivity and selectivity. A homogeneous fluorescence energy transfer assay in the competitive mode based on the binding of biotin and streptavidin was developed as an on-line post-column reaction detection system. The labels used for energy transfer were R-Phycoerythrin conjugated to biotin and Cyanine 5 labeled with streptavidin. The energy transfer peak was measured at 670 nm and excitation was achieved using the 488 nm line of an argon ion laser. The biotin concentration in plasma ultrafiltrate ranged from 0.024 to 6.12 ng/mL (n = 6). The precision of the two controls, 0.24 and 2. 44 ng/mL, was found to be 18.70% and 9.92% relative standard deviation respectively. Accuracy was 10.47% and 1.95% difference from spiked, respectively (n = 6). The limit of detection was 21.70 pg/mL (8.90 x 10(-11)M) calculated based on a factor of 2x the standard deviation of the blank (n = 6). The correlation coefficient for the calibration curve was found to be 0.9995. Recovery from plasma ultrafiltrate at 2.44 ng/mL was 103.40% (n = 6). Detection selectivity was indicated by the absence of background fluorescence in six different plasma samples collected from six individual donors. Endogenous levels were detected in two of the six pools of plasma ultrafiltrates.