The Quantitation of GST in Biological Fluids

For the quantitation of the GST both activity and immunoassay approaches have been tried. The substrate that is most commonly employed is CDNB as it is a substrate for the majority of the GST isoenzymes in humans, the rat, and other species (C13). There are considerable differences in the specific activity of the various subunits to CDNB and therefore interpretation of such measurements is difficult if more than one isoenzyme is present. In the rat, l,2-dichloro-4-nitrobenzene and bromosulfophthalein have been used but these substrates are active only with the rat Yb2 subunit and essentially display no activity with the human GST (B36). 

At present the use of activity measurements to quantitate plasma, serum, or urinary levels of the GST are inadequate. With CDNB it is difficult to obtain sufficient sensitivity to allow the measurement of the levels in normal subjects (A4). In addition many drugs and endogenous substances may inhibit the activity to values that lie within the reference range. For example bile salts and bilirubin inhibit GST activity (H17) and since both of these nonsubstrate ligands are increased in liver disease their accumulation in plasma could theoretically suppress GST activity to within the reference range. An important problem with GST activity measurements concerns the ubiquitous nature of the GST since poor organ specificity will result unless specific isoenzymes are measured. For example, platelets, erythrocytes, and white cells contain high levels of the isoenzyme and these cells may release their GST into plasma prior to separation of the blood sample (G4, H52, L12, M8, Rll, S43). With the substrates that are available to date, the activity measurements are inadequate for clinical use. 

A number of immunoassay approaches have been used for the GST. Western blotting provides a semiquantitative method for studying the GST (H37) but it may be necessary to concentrate the protein by, for example, affinity chromatography prior to blotting. Blotting methods are too insensitive for use in plasma but they are suitable for semiquantitative measurement of GST in tissue. 

Radioimmunoassay (RIA) is the method most commonly used for the measurement of plasma and urinary concentrations of GST. The first RIA described for the GST was for rat ligandin (GST YaYa) (B6) and in 1978 the first assay for human ligandin was described (T13). The assay for human ligandin lacked sensitivity as did the RIA for human ligandin described in 1983 by Sherman et al. (H18, S20). In 1984 we described the first RIA methods that could differentiate the B1 and B2 subunits with a high degree of sensitivity and specificity as shown in Table 9 (B16). These assays could detect both the upper and the lower reference limits for plasma measurements. When animals were immunized with the homodimer B,B or B2B2, the cross-reactivity that the antisera had with the heterodimer B,B2 was quite variable. Because the proportion of the B,B2 heterodimer in liver can vary markedly (H63), it is important to choose antisera that will react only with one of the subunits in the heterodimer. The antisera we used for our subsequent clinical studies on plasma Bj and Bj subunit measurements in plasma or serum exhibited approximately 50% cross-reactivity with the B,B2 heterodimer and thus it was possible to measure total B, or B2 subunits rather than the total subunits in the homodimer and a variable proportion of the same subunit in the homodimer. 

We have also described a specific RIA for the pi-class enzymes that is sensitive enough to detect plasma levels as shown in Table 9 (H52). Other assays for human pi-class GST have also been described using both polyclonal and monoclonal 

---