Key points are not available for this paper at this time.
In humans, allantoin is formed by nonenzymatic oxidation of urate; it may, therefore, be useful in assessing oxidative stress(1)(2). Most published methods involve separate analysis of urate and allantoin and require extraction, hydrolysis, and derivatization procedures (1)(2)(3)(4)(5)(6). The primary aim of this study was to evaluate a slightly modified version of an HPLC assay described by Lux et al. (7) for the simultaneous measurement of urate and allantoin. A secondary aim was to explore the clinical utility of allantoin as a biomarker of oxidative stress, the hypothesis being that in disease associated with increased oxidative stress, allantoin increases because of an increased “oxidative turnover” of urate. The final aim of the study was to investigate the effect of age on urate and allantoin concentrations. Allantoin and uric acid were from Sigma; 1-heptanesulfonic acid, sodium salt monohydrate was from Sigma-Aldrich; potassium dihydrogen phosphate was from Merck; sodium hydroxide was from Riedel-de Haen; orthophosphoric acid was from BDH, and Moni-Trol Level 1 Chemistry Control Serum was from Dade International. MilliQ water (Millipore ultra-pure water system; Millipore) was used for preparation of all solutions. Aqueous stock solutions of allantoin (1000 μmol/L) and urate (2000 μmol/L) were prepared and stored at 4 °C. Because uric acid (urate) is more soluble at alkaline pH, sodium hydroxide (1 mol/L) was added dropwise until the pH was ∼9.0; at this pH, all urate was dissolved. Calibrators (10–100 μmol/L for allantoin; 50–1000 μmol/L for urate) were prepared in mobile phase from stock solutions: 25 μL of each calibrator was mixed with 25 μL of Moni-Trol control serum and 75 μL of mobile phase. Ultrafiltrates (see below) of diluted calibrators were used to construct daily calibration curves. For precision studies, we used 1-mL aliquots of pooled heparinized plasma with or without added allantoin (25 μL of stock solution) and urate (250 μL of stock solution) to prepare control samples. For sample preparation, we vortex-mixed 25.0 μL of sample or control with 100 μL of mobile phase, transferred the mixture into a filter unit (Millipore Ultrafree-MC 30 000 NMWL polysulfone-membrane filter unit; Millipore) that had been prewashed twice before use with 300 μL of water to remove the humectant (glycerol), and centrifuged the mixture in a MSE Micro Centaur (MSE Scientific Instruments) at 2500g for 10 min to remove protein and other molecules of Mr >30 000. Ultrafiltrate (20 μL) was injected into the HPLC system, which comprised an isocratic pump (ISCO model 2350 pump with a 20-μL looped Valco manual injector; ISCO), a variable wavelength absorbance detector (ISCO model V4 detector with 5-mm flow cell path), a cartridge guard column (Spherisorb C18, 5 μm, 10 × 4.6 mm i.d. cartridge; ISCO), and a reversed-phase analytical column (ISCO C18, 5 μm, 250 × 4.6 mm i.d.). The mobile phase was aqueous 5 mmol/L potassium dihydrogen phosphate containing 5 mmol/L 1-heptanesulfonic acid (ion-pairing reagent) and adjusted to pH 3.1 using orthophosphoric acid. The flow rate was 1.0 mL/min, and detection was at 210 nm. The peak heights of allantoin and urate were measured manually, using chromatograms recorded by a chart recorder fitted within the detector. By plotting peak height against calibrator concentration, we constructed calibration curves for allantoin and urate. We calculated the concentration (μmol/L) of allantoin or urate in each sample, using peak height over the slope of the calibration curve. The purities of the compounds of interest were not assessed, but were assumed on the basis of previously published data (7). Because urate reportedly is less stable at alkaline pH (2), the stability of the stock urate calibrator (pH 9.4) was assessed. To check whether membrane filtration caused loss of analyte, fresh fasting, heparinized plasma was analyzed with and without filtration. Linearity was assessed by repeated measurements of Moni-Trol control at various concentrations. Recovery was assessed by the addition of allantoin (25 μmol/L) and urate (250 μmol/L) to pooled plasma. A signal-to-noise ratio of 3:1 was used to determine detection limits. This study was approved by the Ethics Subcommittee of the Hong Kong Polytechnic University, and all procedures involving human subjects complied with the Declaration of Helsinki, as revised in 1996. Fasting heparinized plasma samples were obtained from 40 apparently healthy volunteers 23 men, ages 20–55 years; mean (SD), 30.3 (11.9) years; and 17 age-matched women and 64 subjects with non-insulin-dependent diabetes mellitus NIDDM; 27 men, ages 32–86 years, mean (SD), 63.4 (14.1) years; and 37 age-matched women. The NIDDM subjects had been assessed clinically by the treating physician as suffering from early peripheral vascular disease (PVD). Heparin was the preferred anticoagulant because the elution peak of allantoin was completely masked when EDTA-treated or citrated plasma was used (results not shown) This method was essentially that of Lux et al. (7), modified as follows: plasma was prediluted in the mobile phase; the sample size was reduced to 25 μL; samples were ultrafiltered for 10 min instead of 60 min; the mobile phase was made slightly more acidic, which increased the retention time of urate slightly the retention time of allantoin was not affected (results not shown); the column used was longer, which improved the resolution of allantoin; and the flow rate was slower, which lowered pump pressure, but optimized resolution of the analytes of interest and improved precision. The method showed clear separation of allantoin and urate, with retention times of 3.0 and 9.5 min, respectively (Fig. 1). The detection limit for allantoin and urate in mobile phase was 20 pmol (equivalent to a plasma concentration of 5 μmol/L). The within- (n = 9) and between-day (n = 6) CVs for allantoin (15–25 μmol/L) were 0.61). When compared with the results for healthy subjects, urate was significantly higher (P <0.005) in NIDDM women, but not NIDDM men. In NIDDM subjects, a significant direct correlation was seen between age and urate concentration (r = 0.483; P <0.001) and between age and allantoin concentration (r = 0.439; P <0.001). Age could account for only ∼20% of the variation in these variables, however, and no significant correlation was seen between age and allantoin:urate ratio (r = −0.133). This implies that the difference in the pattern of results between healthy and NIDDM subjects studied was not caused solely by the difference in age between the groups. Plasma allantoin has been reported to increase in conditions associated with increased oxidative stress and to decrease after antioxidant supplementation(2)(5)(8)(9). However, few studies have been performed to date, methods have not been standardized, and subject numbers have generally been small (see Table 1) (2)(3)(5)(6)(7)(8)(9). No sex- or age-related effects had been investigated prior to the current study, and to our knowledge, no data had been reported on NIDDM subjects. Results of this study showed that NIDDM subjects with early PVD have increased allantoin concentrations. NIDDM is associated with lowered antioxidant status and increased oxidative stress, and this may play a role in the development of diabetic complications (10)(11)(12)(13)(14)(15)(16)(17)(18)(19). The increased plasma allantoin concentrations in the NIDDM subjects studied here support the concept of using plasma allantoin as a biomarker of oxidative stress (2)(3). However, although 20% of NIDDM subjects showed a marked increase in allantoin (more than 2 SD greater than the mean of the control group), there was considerable overlap between the groups, and further study of allantoin in relation to oxidative stress clearly is needed. Urate has been reported to be an important physiological antioxidant(20)(21)(22), and it makes a large contribution to the measured “total antioxidant capacity” of plasma(23)(24). Urate is an independent risk factor for coronary heart disease (25)(26), however, and urate concentrations have been reported to correlate inversely with vitamin E concentrations and directly with lipoperoxides (as TBARS)(27). In this study women, but not men, with NIDDM had increased urate concentrations. This is an interesting finding in view of the greater increase in coronary heart disease risk found in NIDDM women compared with men (28). In conclusion, the relatively straightforward HPLC method evaluated here is suitable for routine use. This method will facilitate further clinical evaluation of allantoin, urate, and allantoin:urate ratios in fasting plasma and other biological fluids as biomarkers of oxidative stress. We thank The Hong Kong Polytechnic University for financial support for this work. We also thank the Hong Kong Research Grants Council for funding (Earmarked Research Grant CUHK 425/95M) the study of diabetic patients with early PVD, some of whose results are reported here.
Benzie et al. (1999) studied this question.