Ultrasensitive Direct Fluorescent Immunoassay for Thyroid Stimulating Hormone

^a author for correspondence: fax 410-740-2985, e-mail martek2001{at}aol.com

Phycobilisomes are photosynthetic antennae complexes of red algae and cyanobacteria (1)(2)(3) . They have been chemically cross-linked in such a way that they remain soluble and stable (4) . These stabilized phycobilisomes (PBXL^TM dyes) have large complex weights (between 1.0 x 10⁷ and 1.5 x 10⁷ Da) and Stokes shifts. They contain a large number of chromophores coordinated to efficiently transfer energy down an energy gradient and emit between 662 and 666 nm. The PBXL-1 dye, used in the thyroid-stimulating hormone (TSH) model, contains B-phycoerythrin, R-phycocyanin, and allophycocyanin as its component phycobiliproteins. Each PBXL supramolecular complex can deliver up to 1400 chromophores per binding event without indirect signal generation steps, signal amplification, or enzyme substrates. PBXL dyes provide physical amplification of signal, enabling ultrasensitive direct fluorescent immunodetection of such clinically relevant analytes as TSH.

TSH was used as a model system for a microplate immunoassay because of its clinical importance and high sensitivity requirements. The detection limit of the PBXL based assay was 0.01 mIU/L (6.2 x 10^-14 mol/L).

We used the following reagents: plate-coating buffer containing 100 mmol/L sodium phosphate (pH 7.4), 150 mmol/L sodium chloride, and 0.5 g/L sodium azide; wash buffer containing 100 mmol/L sodium phosphate (pH 7.4), 150 mmol/L sodium chloride, 0.5 g/L sodium azide, and 0.5 g/L Tween 20; blocking buffer containing 100 mmol/L sodium phosphate (pH 7.4), 150 mmol/L sodium chloride, 0.5 g/L sodium azide, 0.5 g/L Tween 20, and 10.0 g/L bovine serum albumin; assay buffer containing 100 mmol/L sodium phosphate (pH 8.4), 150 mmol/L sodium chloride, 0.5 g/L sodium azide, 0.5 g/L Tween 20, and 10.0 g/L bovine serum albumin; and reagent dilution buffer containing 100 mmol/L sodium phosphate (pH 8.4), 150 mmol/L sodium chloride, 0.5 g/L sodium azide, and 10.0 g/L bovine serum albumin.

The capture antibody was anti-TSH-ß monoclonal, cat. no. 10-T25, clone no. M94204 (Fitzgerald Industries International). The tag antibody, MAB131, anti-TSH-ß monoclonal, lot no. 606272 (AbProbe International), was conjugated to PBXL-1 via SATA/sulfo-SMCC heterobifunctional cross-linking chemistry (Pierce) at a 18:1 offered molar ratio.

To make calibrators, purified TSH (Scripps Laboratories, Second IRP for TSH 80/558) was added to a matrix of 60.0 g/L bovine serum albumin, 100 mmol/L sodium phosphate (pH 7.4), and 150 mmol/L NaCl.

Each well of a black 96-well plate (Dynex Technologies) was coated with 150 µL of 100 mg/L capture antibody. The plates were covered and allowed to sit at room temperature overnight. The plate-coating buffer was aspirated, and the plates were washed three times with 350 µL/well of wash buffer. The plates were blocked with 350 µL/well blocking buffer at 37 °C for 2 h. The blocking buffer was aspirated, and the plates were washed four times with 350 µL/well wash buffer. The coated plates were then used immediately for the TSH assay.

In the assay, we added 75 µL of calibrator and 75 µL of assay buffer to each well. The plates were covered and incubated for 90 min at 37 °C. The plates were washed three times with 350 µL/well wash buffer. Fifty microliters of PBXL-1:Anti-TSH conjugate (100 mg/L in reagent dilution buffer) and 100 µL of assay buffer were added to each well. The plates were covered and incubated for 90 min at 37 °C. The plates were then washed three times with 350 µL/well wash buffer. One hundred microliters of plate-coating buffer was added to each well. The fluorescence of each well was determined on a Fluorolite 1000 plate reader (Dynex Technologies) at 10 V using 550 nm excitation and 660 nm emission filters.

The five-point calibration curve (0.0, 0.05, 1.0, 5.0, and 10.0 mIU/L) used a weighted logit fit analysis (Fig. 1 ).

View larger version (17K): [in this window] [in a new window]   Figure 1. Five-point calibration curve for TSH PBXL-1 immunoassay.

The linearity and limit of detection were determined by assaying dilutions of the 10 mIU/L calibrator in the zero calibrator matrix (10.0, 5.0, 2.5, 1.0, 0.5, 0.2, 0.1, 0.05, 0.01, 0.005, and 0.0 mIU/L) in replicates of six. Concentrations were determined from a five-point calibration curve. Linear regression analysis of all points demonstrated a slope of 0.983 (1.010–0.956, upper and lower 95% confidence interval), a y-intercept of 0.019 mIU/L (0.110 to -0.071 mIU/L, upper and lower 95% confidence interval), standard error of the estimate of 0.334 mIU/L, and r² of 0.996. Quadratic regression analysis for curvature demonstrated a P value of 0.06, indicating lack of evidence for significance of the curvature of the line. The analytical limit of detection was 0.01 mIU/L, as calculated from two times the SD of 20 replicates of the 0 mIU/L calibrator. Similarly, 0.01 mIU/L was the lowest concentration dilution that demonstrated a statistically significant difference (P 0.05) from the zero calibrator, as determined by a non-paired Student's t-test analysis.

The within-run imprecision (CV) for replicates of 20 was 6.4% and 4.5% at concentrations of 0.56 and 7.2 mIU/L, respectively.

The dynamic linear assay range was 0.01–10.0 mIU/L.

A 0.01 mIU/L analytical detection limit (6.2 x 10^-14 mol/L) has been possible until now only with indirect detection technologies that involve enzymatic signal generation (e.g., chemiluminescence or chemifluorescence). PBXL pigments allowed this detection limit within the limitations of a manual assay format without enzymatic signal generation steps. It is anticipated that both detection limit and precision would improve if the assay were formatted on automated instrumentation developed for the features of the PBXL dyes.

The cost per test of PBXL pigments is competitive with traditional detection systems, and the stability of the PBXL conjugates is in excess of 1 year. PBXL technology has the combined advantages of ease of use, low cost, stability, and high sensitivity to set a new standard of performance as demonstrated by the TSH immunoassay model. This new technology has considerable potential for enhancing sensitivity in applications such as assay miniaturization, blotting technologies, and DNA arrays.

Footnotes

Martek Biosciences Corporation, 6480 Dobbin Rd., Columbia, MD 21045

References

1. Gagliano AG, Hoarau J, Breton J, Geacintov NE. Orientation of pigments in phycobilisomes of Porphyridium sp. Lewin. A linear dichroism study utilizing electric and gel orientation methods. Biochim Biophys Acta 1985;808:455-463.
2. Gantt E, Lipschultz CA. Phycobilisomes of Porphyridium cruentem. J Cell Biol 1972;54:313-324. [Abstract/Free Full Text]
3. Yamanaka G, Glazer AN, Williams RC. Molecular architecture of a light-harvesting antenna. Comparison of wild type and mutant Synechococcus 6301 phycobilisomes. J Biol Chem 1980;255:11004-11010. [Abstract/Free Full Text]
4. Cubicciotti RC, . assignee. Phycobilisomes, derivatives, and uses therefore. US Patent 5,695,990 1997;.