Summary

Cayman Chemical Company has developed a fluorescence polarization immunoassay to quantify Prostaglandin E₂ (PGE₂). This assay is ideal for high-throughput applications because it takes place in a homogeneous environment (i.e., solution) and requires short incubation times (approximately one hour). The following is a description of how this assay was validated, how robustness was quantified, and how it performs in a typical application.

Standard Curve Validation

The antibody and tracer (i.e., fluorescein labeled PGE₂) were titered to optimize dynamic range and limit of detection. Dynamic range is the difference in mP (ΔmP) between the highest and lowest standard. Based on previously published assays, this difference should be ≥100. The assay described here has a ΔmP of 175. Limit of detection is the lowest concentration of analyte reliably quantified (i.e., CV<20%). Currently, the limit of detection is approximately 156 pg/mL.

Standard curves were then constructed by serially diluting PGE₂ in FP buffer (PBS + 0.1 mg/mL BSA) between 20 ng/mL and 156 pg/mL. Millipees were determined at each concentration for every curve. Six curves were run each day for six days. For each curve, the mP data was transformed using a logit function, y = ln(x/(1-x)), and plotted versus concentration. The data was then fit using a log-regression algorithm in Microsoft Excel. The concentration of each standard was then recalculated using the regression parameters. All values corresponding to a particular standard were averaged and the coefficient-of-variance determined. Figure 1 is a typical standard curve overlaid with the %CV curve. Also, the development of the standard curve was examined over time. Figure 2 show the development of a curve from 15 minutes to 18 hours. At low concentrations, the curve develops immediately. At higher concentrations, the development is within validation limits at 60 minutes, and does not change significantly even at 18 hours.

Figure 1. PGE₂ Standard Curve with %CV of Each Point

Figure 2. Development of the Standard Curve from 15 minutes to 18 hours

Z′-Factor

The Z′-Factor is a statistic designed to reflect both the assay signal dynamic range and the variation associated with the signal measurements. Prior to the definition of this coefficient, the quality of assays were determined based on their signal-to-noise (S/N) or signal-to-background (S/B) ratios. Both of these parameters, however, suffer because they do not take into account the variability in the sample and background measurements and the signal dynamic range. The Z′-Factor accomplishes this, and, because it is dimensionless, can be used to compare similar assays. The equation and the meaning of possible values are summarized in table 1.

To determine the Z′-Factor for the PGE₂-FPIA, five 96 well plates were run each day over the course of three days. Each plate had 48 negative control wells and 48 positive control wells. The negative control wells contained only FP buffer, and the positive control wells contained FP buffer spiked with 2000 pg/mL PGE₂. The mP values were determined for all wells in every plate, and %B/B₀ was calculated from this data. The average %B/B₀ for the negative controls was 100.04, and the average for the positive controls was 53.98. The Z′-Factor had a value of 0.88, indicating a very robust assay. Figure 3 is a plot of the data used to determine the Z′-Factor along with lines delineating the ±3σ limits of the positive and negative controls.

Figure 3. Z′-Factor Data with ±3σ Lines

PGE₂-FPIA in Cell Culture

RAW 264.7 cells were grown in DMEM containing 10% FBS until 80% confluent. The cells were then stimulated with 1 µg/mL LPS, 1 µg/mL LPS with 500 µM L-NMMA, or vehicle, for 18 hr. At that point, the cells were washed once with either pre-warmed HBSS or PBS, then incubated for 20 minutes at 37°C with either warm HBSS or PBS containing 0.1% BSA and 20 µM arachidonic acid. At the end of the incubation period, the supernatants were collected and immediately frozen at -80°C.

When assayed, the supernatant was thawed at room temperature without assistance. Three separate samples were aliqouted from each supernatant. The first aliquot (A) was left as a blank, the second (B) was spiked with 10 ng/mL PGE₂, and the third was spiked with 30 ng/mL PGE₂. All aliqouts were diluted 1:8 (v/v) with FP buffer and assayed. To control for any interferences, two standard curves were used. One prepared with PBS and the other with HBSS. The mP values for the standard curves were transformed using the logit function, and the regression line was calculated. The concentrations of the samples were then calculated from the regression parameters. Table two summarizes the results of this experiment.

Discussion

The above outlines the validation of a PGE₂-FPIA. The variance of the standard curve is below similar assays (e.g., EIA) available for the quantitation of PGE₂. The robustness of this assay for high-throughput screening was established by determining the Z′-Factor (0.88). Finally, the utility of this assay in a real-world application was demonstrated.