Analytical methods impact estimates of trichloroethylene's glutathione conjugation and risk assessment.

ABSTRACT

The glutathione (GSH) conjugates, S-(1,2-dichlorovinyl)-glutathione (DCVG) and S-(1,2-dichlorovinyl)-L-cysteine (DCVC), have been implicated in kidney toxicity and kidney cancer from trichloroethylene (TCE) exposure. Considerable differences in blood and tissue levels of DCVG and DCVC have been reported, depending on whether HPLC/UV (High Performance Liquid Chromatography-Ultraviolet) or HPLC/MS (HPLC-Mass Spectrometry) was used. A side-by-side comparison of analytical results with HPLC/UV and HPLC/MS/MS (High Performance Liquid Chromatography-Tandem Mass Spectrometry) detection was undertaken to quantitatively compare estimates for DCVG and DCVG using rat and human tissues. For the HPLC method, DCVG and DCVC were initially derivatized with fluorodinitrobenzene (DNP). The results from the HPLC/UV method showed that derivatized-DCVC eluted at the solvent front and could not be quantified. Derivatized-DCVG, however, was quantified but significant interference was observed in all four control tissues (rat blood, liver, kidney; and human blood), resulting in average spike recoveries of 222-22,990%. In contrast, direct analysis of spiked tissues by HPLC/MS/MS resulted in recoveries of 82-127% and 89-117% for DCVG and DCVC, respectively. These differences in analytical results were further confirmed in tissues from TCE-treated rats, e.g., DCVG levels in rat liver were 18,000 times higher by HPLC/UV as compared to HPLC/MS/MS. Fraction collection of the derivatized-DCVG peak (obtained with the HPLC-UV method), followed by peak identification via an HPLC/UV/Q-TOF/MS/MS method, identified DNP-derivatized endogenous glutamate as the primary interfering substance that contributed to and exaggerated recoveries of DCVG. Thus, estimates of DCVG based on the HPLC/UV methods are not reliable; they will over-estimate the formation of the GSH conjugates of TCE and will artifactually exaggerate the potential cancer risk in humans from TCE exposure. Therefore, it is recommended that any characterization of cancer risks from TCE exposure attributable to the GSH conjugates of TCE rely on results obtained with the more specific and reliable HPLC/MS/MS method.