Determination of four oxidative stress biomarkers in human urine using solid-phase extraction coupled with ultra performance liquid chromatography-tandem mass spectrometry

Oxidative stress biomarkers are measurable biological indicators that reflect the balance between the production of reactive oxygen species (ROS) and the body's ability to neutralize them using antioxidants. Elevated oxidative stress is associated with a number of health effects. Herein, we report the development of a comprehensive and sensitive method for quantifying four typical oxidative stress biomarkers in human urine using solid-phase extraction (SPE) in conjunction with ultra performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS). The quantified biomarkers include L,L-dityrosine (diY), 8-hydroxy-2'-deoxyguanosine (8-OHdG), 8-hydroxyguanosine (8-OHG), and 4-hydroxynonenal mercapturic acid (HNEMA), which are markers of oxidative-stress-related damage in proteins, DNA, RNA, and lipids, respectively. To that end, we systematically optimized the MS parameters, SPE cartridge, and elution conditions of the method. Briefly, 0.2 mL of a urine sample was mixed with 0.8 mL of pure water, after which an internal-standard mixture was added. The four target analytes were enriched and purified using an HLB SPE cartridge. The diY and the other three compounds were eluted with 2% (volume fraction) methanol aqueous solution and methanol, respectively. The two groups of eluates containing different target analytes were separately injected onto an Acquity UPLC HSS T3 column (100 mm×2.1 mm, 1.8 μm) and gradient eluted using 0.05% (v/v) acetic acid aqueous solution and methanol. The target analytes were identified using both negative and positive electrospray ionization (ESI^- and ESI⁺) and multiple reaction monitoring (MRM) modes, and quantified using stable-isotope-labeled internal standards. The four typical oxidative-stress biomarkers exhibited good linearities within the mass concentration range of 0.01-100 μg/L, with correlation coefficients ≥0.9998, and limits of detection (LODs) and limits of quantification (LOQs) of 7-18 and 22-60 ng/L, respectively. The spiked recoveries of the target analytes at three levels (5, 10 and 50 μg/L) were 103.0%-105.6%(8-OHdG), 100.8%-104.2%(8-OHG), 97.2%-100.2%(diY) and 96.9%-106.0%(HNEMA), with intra-day precisions of between 1.6% and 5.2%. Moderate-to-strong matrix effects of between 42% and 137% were observed for each target analyte. The target compounds exhibited weak matrix effects of 99%-102% (8-OHdG), 97%-98% (8-OHG), 97%-106% (diY), and 94%-110% (HNEMA) after adjustment using the stable-isotope-labeled internal-standard method. The developed method was used to determine the abovementioned four typical oxidative stress biomarkers in 40 urine samples. All target compounds were detected in human urine at rates of 100%, with mass concentrations of 0.52-14.40 μg/L, 2.75-38.15 μg/L, 8.92-82.28 μg/L, and 1.74-575.29 μg/L recorded for 8-OHdG, 8-OHG, diY, and HNEMA, respectively, along with median values of 2.89, 12.36, 37.66, and 96.92 μg/L, respectively. The developed method is simple to operate, highly sensitive, and is very precise and accurate; consequently, it is suitable for determining the abovementioned four typical oxidative stress biomarkers in human urine.