Final Report

Major Goals and Objectives:  The main goal of the research was to conduct preliminary research that would lead to the development of a rapid assay for the analysis of PCBs in fish tissue, in order to offer options to the expensive and time-consuming FDA method. This method would be less expensive, less time consuming and produce less waste then the method currently being used.

The major objectives of the research were:

1. To improve the extraction and clean-up methods before using a commercial enzyme-linked immunosorbent assay (ELISA) for the analysis of PCBs in fish tissue. In addition, an aim of the research is to increase the sensitivity and the selectivity of the ELISA.
2. To demonstrate the effectiveness of this method for the analysis of fish tissue collected during an annual State-wide survey.
3. To compare total PCB concentrations determined using gas-chromatography/electron capture detection (GC/ECD) with those determined using ELISA.
4. To apply the extraction and cleanup procedure for another ELISA to measure the toxicity equivalence quotient (TEQ) in fish.

Summary of Progress towards Goals and Objectives:  Standard curves for Aroclor 1254, developed using fortified catfish tissue, gave a linear range for the two standard curves (0.05 to 0.5 ppm and 0.5 to 5.0 ppm) and correlation coefficients of 0.97 and 0.98, respectively. Eleven fish samples that were previously found to contain Aroclor 1254 and 1260 at levels above the detection limit of 50 ppb, were analyzed for their total PCB content using ELISA. The values obtained from the ELISA were comparable to those from the standard FDA GC/ECD method.

Accomplishments:  From the data obtained thus far, it appears that the clean-up, extraction and the ELISA are more economical, efficient and produce less waste than the conventional FDA method. This method may allow State agencies to expand their sample numbers with increased throughput. This method can also be used to develop fish consumption advisories that more accurately articulate health risks from the consumption of contaminated fish.

Keywords: PCBs, fish, consumption advisory, ELISA

Narrative Report:  Catfish tissue was obtained from a local grocery store and ground into a homogenous mixture. The final fat content of the catfish homogenate was adjusted to 5.2% by adding an appropriate amount of belly flap tissue. Fish tissue (1.0000g) was mixed with anhydrous sodium sulfate and allowed to dry for several hours. The PCB in the fish sample was extracted through a sulfuric acid-silica gel column with 20 ml hexane. An aliquot of the analyte was exchanged into 50% (v/v) methanol and analyzed using ELISA.

Standard curves for PCBs in fish were obtained in the ranges of 0.05 to 0.5 ppm and 0.5 to 5.0 ppm of Aroclor 1254. The minimum and the most appropriate volume of hexane used for elution of PCBs, was found to be 20 mL. The dilution factors for the two curves were calculated prior to each sample analysis and were done such that the upper concentration of each standard curve coincided with the maximum working range of the kit (i.e., 5 ppb). Therefore, the dilution factors for the two standard curves were 100 and 1000. Standard curves for the range of 0.05 to 0.5 ppm and 0.5 to 5 ppm were constructed by plotting Logit B/Bo (relative concentration) against Ln (concentration). The Aroclor 1254 standard curve ranging from 0.05 to 0.5 ppm gave a correlation coefficient of 0.97. The curve ranging from 0.5 to 5.0 ppm gave a correlation coefficient of 0.98.

Eleven contaminated fish samples, collected during an Indiana fish survey, were analyzed using ELISA. All the samples tested had detectable PCB residues. The levels detected ranged up to 519 ppb. These samples were previously tested using the FDA GC/ECD method and found to have levels ranging up to 570 ppb. A t-test was carried out for data collected from the FDA GC/ECD and the ELISA methods. No significant differences were found between these data sets (degrees of freedom = 9, t(9)_0.05=1.833, t_s=2.19, reject H_o at p<0.05).

In the experiments carried out, the extract solvents used for the ELISA were methanol and water in the ratio of 1:1. The tests were also done randomly to remove any bias while carrying out the extraction and the ELISA. The standard curve ranging from 0.05 to 0.5 ppm Aroclor 1254 was repeated using the methanol and the ELISA zero standard as the diluent for the sample extract. The ELISA temperature was controlled in a water bath at a temperature of 23C. The standard curve ranging from 0.05 to 0.5 ppm Aroclor 1254 gave a correlation coefficient of 0.976.

During the initial stages of experimentation several problems were resolved with regard to linearity. For that purpose, a pilot study was conducted wherein ¹⁴C-PCB 153 was used to track the fate of PCB through the column. The data indicated that about 95% PCB was being recovered from the column and that the losses of PCB could not be owed to loss within the column but were probably the result using of water as a solvent (low PCB solubility).  However, further work needs to be carried out to determine the selectivity of the antibodies for congeners present in Aroclor 1242, 1248 and 1260. The extraction and cleanup method developed here will next be used in conjunction with another ELISA to determine the TEQ for various fish samples.

Graduate Students Supported:

 Ping Wan, M.S. student
Joanne Lasrado, Ph.D. student

Table 1: Comparison of ELISA data with GC/ECD data for total PCB in contaminated fish samples.