Analysis of endogenous glutathione-adducts and their metabolites

Abstract

The ability to conduct validated analyses of glutathione (GSH)-adducts and their metabolites is critically important in order to establish whether they play a role in cellular biochemical or pathophysiological processes. The use of stable isotope dilution (SID) methodology in combination with liquid chromatography-tandem mass spectrometry (LC-MS/MS) provides the highest bioanalytical specificity possible for such analyses. Quantitative studies normally require the high sensitivity that can be obtained by the use of multiple reaction monitoring (MRM)/MS rather than the much less sensitive but more specific full scanning methodology. The method employs a parent ion corresponding to the intact molecule together with a prominent product ion that obtained by collision induced dissociation. Using SID LC-MRM/MS, analytes must have the same relative LC retention time to the heavy isotope internal standard established during the validation procedure, the correct parent ion and the correct product ion. This level of specificity cannot be attained with any other bioanalytical technique employed for biomarker analysis. This review will describe the application of SID LC-MR/MS methodology for the analysis of GSH-adducts and their metabolites. It will also discuss potential future directions for the use of this methodology for rigorous determination of their utility as disease and exposure biomarkers.

Figure 1

Formation of GSH-adducts and small molecule biomarkers.

Figure 2

SID LC-MRM/MS chromatograms of lower limit of quantification samples with 0.0005 mM GSSG and 0.005 mM GSH. MRM chromatograms for GSABD (m/z 505.3 → 376.2), GSSG (m/z 613.2 → 355.2), [¹³C₂, ¹⁵N₁]-GSABD (m/z 508.3 → 379.2) and [¹³C₄, ¹⁵N₂]-GSSG (m/z 619.2 → 361.2). Reprinted with permission from Zhu et al. (2008).

Figure 3

Formation of eicosanoids and eicosanoid-derived GSH-adducts. Abbreviations: 5-HEDH, 5-hydroxyeicosanoid dehydrogenase; EH, epoxide hydrolase; FLAP, 5- lipoxygenase activating protein; FOG-7, 5-oxo-eicosatetraenoic acid GSH-adduct, LTAH, leukotriene A₄ hydrolase; LTAS, leukotriene A₄ synthase; PGS, prostaglandin synthase; TX, thromboxane; TXS, thromboxane synthase.

Figure 4

Lipid hydroperoxide-derived bifunctional electrophiles that can form GSH-adducts.

Figure 5

Metabolism of HNE to GSH-adducts and MA derivatives. AD, aldehyde dehydrogenase.

Figure 6

HNE-GSH determination (MRM m/z 464.3 → 308.1) in rat liver tissue (a) of a vehicle-treated control rat and (b) 5 h after a single ip dose (15 mg/kg) of Fe(III)NTA. Reprinted with permission from Völkel et al. (2005).

Figure 7

Formation of TOG from a lipid hydroperoxide [15(S)-HPETE].

Figure 8

Quantitative analysis of TOG and HNE-GSH-adducts after adding 10 μM t-butyl-hydroperoxide and 500 μM Fe^II to EA.hy 926 endothelial cells. The upper chromatogram shows the MRM signal for endogenously generated TOG m/z 426 (MH⁺) → m/z 280. The center chromatogram shows the MRM signal for the [²H₃]-TOG internal standard m/z 429 (MH⁺) → m/z 283 and the lower channel shows the MRM signal for endogenously generated HNE-GSH-adducts and ONO-GSH-adducts m/z 464 (MH⁺) → m/z 308 (MH⁺-C₉H₁₆O₂). The concentration of intracellular TOG and the HNE-GSH-adduct corresponded to 8.6 and 0.5 μM, respectively, as determined from a standard curve constructed in blank cell lysate buffer. Reprinted with permission from Jian et al. (2007).

Figure 9

LC/MS/MS profile of LTC₄, its metabolites, and selected eicosanoids in mouse peritoneal lavage. Mice were treated with 1 mg zymosan for 2 h, and eicosanoids analyzed by SID LC-MRM/MS, monitoring the specific m/z transitions indicated. Retention times of deuterated internal standards (coincided with their cognate metabolites. Reprinted with permission from Zarini et al. (2009).

Figure 10

Evaluation of ion suppression on response of LTE₄ and LTE₄-d3 in urine and water. Injection experiments: extracted ion chromatogram (EIC) of MRM transitions for LTE₄-d3 (m/z 443.2 → 304.3) spiked at 200 pg/mL in a human urine sample (A) and water (B), and MRM transition of endogenous LTE₄ (m/z 440.2 → 301.3) in a human urine sample (C, hatched lines). Data demonstrate that the response of internal standard is not affected by ion suppression. Data was collected in positive electrospray ionization mode. The peak area for urine and water spiked with LTE₄-d3 was 115754 and 107672, respectively. Post-column infusion experiments: LTE₄ was infused and the 440.2 → 301.2 transition was monitored during injection of urine (D) and water (E) samples. Data demonstrate the absence of ion suppression in a urine matrix. Reprinted with permission from Armstrong et al. (2009).