5-N-Carboxyimino-6-aminopyrimidine-2,4(3H)-dione, a novel indicator for hypochlorite formation

Abstract

Uric acid is an adequate and endogenous probe for identifying reactive oxygen or nitrogen species generated in vivo because its oxidation products are specific to reacted reactive oxygen or nitrogen species. Recently, we identified 5-N-carboxyimino-6-N-chloroaminopyrimidine-2,4(3H)-dione as a hypochlorite-specific oxidation product. 5-N-carboxyimino-6-N-chloroaminopyrimidine-2,4(3H)-dione was anticipated to be a biomarker for hypochlorite production in vivo. However, while it was stable in aqueous solution at weak acidic and alkaline pH (6.0–8.0), it was unstable in human plasma. In this study, we found that 5-N-carboxyimino-6-N-chloroaminopyrimidine-2,4(3H)-dione rapidly reacted with thiol compounds such as cysteine and glutathione to yield 5-N-carboxyimino-6-aminopyrimidine-2,4(3H)-dione, which was stable in human plasma unlike 5-N-carboxyimino-6-N-chloroaminopyrimidine-2,4(3H)-dione. 5-N-carboxyimino-6-aminopyrimidine-2,4(3H)-dione was produced upon uric acid degradation during myeloperoxidase-induced uric acid oxidation and lipopolysaccharide-induced pseudo-inflammation in collected 2,4(3H)-dione has potential as a marker for hypochlorite production in vivo.

Introduction

It has been a well-accepted concept that oxidative stress is intimately related to aging and various diseases. Since oxidative stress is caused by reactive oxygen species (ROS), identifying the ROS generated in vivo offers great benefits for understanding oxidative stress and clarifying disease pathology. However, it is difficult to directly detect ROS in vivo because of their instability resulting from their high reactivity. Therefore, it is a reasonable strategy to identify ROS in vivo by monitoring oxidation products specific to each ROS. Thus, we have searched for ROS-specific oxidation products of uric acid (UA). UA is an end metabolite of purine in humans and a ubiquitous water-soluble antioxidant that widely reacts with ROS to produce ROS-specific oxidation products (Fig. 1). We found that parabanic acid (PA) and its hydrolysate, oxaluric acid (OUA), are singlet oxygen (¹O₂)-specific oxidation products of UA.^(1,2)

Fig. 1.

UA oxidative metabolites induced by ROS: AL is formed by free radical-induced UA oxidation. ONOO⁻ reacts with UA to form CA and TU. PA and its hydrolysate OUA are specific to ¹O₂-induced UA oxidation. CCPD is produced by the reaction of UA and ClO⁻ and converted into CAPD by reaction with a thiol group (this study).

Hypochlorous anion (ClO⁻) is one ROS that is intimately related to inflammation. Under inflammatory conditions, myeloperoxidase (MPO) is released from activated neutrophils and catalyzes the reaction of hydrogen peroxide (H₂O₂) and chloride anion (Cl⁻) to form the strong bacteriocide ClO⁻. Recently, we identified 5-N-carboxyimino-6-N-chloroaminopyrimidine-2,4(3H)-dione (CCPD) as a ClO⁻-specific oxidation product.⁽³⁾ CCPD was expected to be an adequate marker for ClO⁻ formation in vivo. However, it was unstable upon addition to isolated human plasma, although it was stable in phosphate buffer solution (pH 7.4). The chloroamino group in the CCPD molecule is reactive toward thiol groups that abundantly reside on proteins in plasma. Therefore, the CCPD was thought to be immediately converted into a secondary product.

In this study, we identified 5-N-carboxyimino-6-aminopyrimidine-2,4(3H)-dione (CAPD) as the thiol-induced CCPD metabolite using an HPLC equipped with a time-of-flight mass spectrometer (LC/TOFMS). CCPD was reacted with cysteine (Cys) or reduced glutathione (GSH) to confirm CAPD formation. As expected, CAPD was formed by the reactions of CCPD with Cys and GSH. CAPD was relatively stable in neutral aqueous solution and human plasma for at least 3 h. CAPD formation was also observed during MPO-induced UA oxidation. CAPD production was investigated during pseudo-inflammation in human blood induced by addition of lipopolysaccharide (LPS). The concentration of CAPD increased after 2 h of LPS addition whereas it did not increase in the absence of LPS. MPO up-regulation in blood was also observed using immune-blotting after addition of LPS. These results indicate that the reaction of UA and ClO⁻ generated from activated neutrophils occurred to form CAPD via CCPD production. Therefore, CAPD is a potential novel marker for ClO⁻ formation in vivo.

Materials and Methods

Chemicals

UA, GSH, Cys, bovine serum albumin (BSA), and other chemicals were purchased from Fujifilm Wako Pure Chemical Co. (Osaka, Japan) and Tokyo Chemical Industry Co., Ltd. and used as received. CCPD was synthesized by reaction of UA and sodium hypochlorite, and purified by HPLC as described previously.⁽³⁾ Pure CCPD was reacted with excess amounts of GSH or Cys to be converted into CAPD. CAPD formation was confirmed by LC/TOFMS as described below. CAPD was isolated and purified by HPLC. An aqueous standard solution of CAPD was prepared and stored under 4°C until use.

Reaction of CCPD with thiol compounds and purification of CAPD

Purified CCPD was dissolved in 40 mM phosphate buffer (pH 7.4). An equivolume solution of Cys (1.0 mM) or GSH (1.0 mM) was immediately added to the CCPD solution and the reaction mixture was analyzed by an optimized LC/TOFMS system. The formed CAPD was isolated and purified using an HPLC system as described below. Additionally, a gradual reaction of CCPD and Cys or GSH was conducted. Next, 10 mM Cys or GSH solution was induced into 20 ml of CCPD (20 μM) solution at a constant rate (0.5 μl/min) using a syringe pump (Harvard apparatus). Changes in concentrations of CCPD and CAPD were determined by an HPLC equipped with a UV detector described below.

Stability of CAPD in aqueous solution and human plasma

CAPD (1.0 μM) was dissolved in 20 ml of phosphate buffer solution (40 mM) adjusted to various pHs (4.0, 5.0, 6.0, 7.0, and 8.0). The solutions were incubated at 37°C for 6 h and changes in the concentration of CAPD were determined by an HPLC equipped with a tandem mass spectrometer (LC/MS/MS) described below. Healthy human blood was collected using heparin as anticoagulant and separated into red blood cells and plasma by centrifugation (2,330 ×g, × 10 min). CAPD (1.0 μM) was added to 5 ml of the separated plasma and the mixture was incubated at 37°C for 6 h. 50 μl of the plasma sample was taken every 1 h and 200 μl of methanol was added to remove contained proteins. After centrifugation (21,880 ×g, × 5 min), 5 μl of supernatant was analyzed by LC/MS/MS.

CAPD formation by MPO-induced UA oxidation

UA was oxidized with ClO⁻ formed from the MPO system. An acetate buffer (50 mM, pH 5.0) containing 40 mM NaCl was prepared. UA (400 μM), BSA (1.0 mg/ml) and MPO (250 mU/ml) were dissolved in the buffer and incubated at 37°C. H₂O₂ aqueous solution (1.14 M) was gradually added to the solution at a constant rate (0.05 μl/min) using a syringe pump. The reaction solution was analyzed every 1 h for 5 h by HPLC-UV as described below.

Pseudo-inflammation in collected human blood induced by LPS

Healthy human blood (approximately 20 ml) was collected using heparin as an anticoagulant and divided in half. LPS (2.5 μg/ml) was added to one half. These blood samples with (+LPS) or without (control) addition of LPS were incubated at 37°C. During incubation, 500 μl samples were taken every 1 h and centrifuged (2,000 ×g, × 5 min). The supernatant was collected as plasma and analyzed by LC/MS/MS and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) as described below.

HPLC-UV analysis

Measurement of changes in concentrations of GSH, cysteine, CCPD, and CAPD was carried out with a reverse phase HPLC equipped with a UV detector (HPLC-UV). The aqueous formic acid (adjusted to pH 3.0) mobile phase was delivered at 1.0 ml/min using an ODS column (TSK-GEL ODS-100Z, 5 μm, 4.6 mm × 150 mm; TOSOH, Tokyo, Japan) for separation. Detection was performed by monitoring the absorption at 210 nm.

LC/TOFMS analysis

To obtain accurate mass-to-charge ratio (m/z) values of CCPD and CAPD, an HPLC equipped with an optimized LC/TOFMS was used. All the acquired m/z values were calibrated with trifluoroacetic acid (TFA) as an internal standard. The aqueous formic acid (pH 3.0) mobile phase was delivered at 1.0 ml/min using a Develosil C30-UG separation column (5 μm, 4.6 mm × 250 mm; Nomura Chemical Co. Ltd, Tokyo, Japan). Negative ionization was carried out at an ionization potential of −2,000 V. Applied voltages to the ring lens, outer orifice, inner orifice, and ion guide were set to −5 V, −20 V, −5 V, and −500 V, respectively.

LC/MS/MS analysis

Measurement of CAPD concentration in plasma was performed by an HPLC equipped with LC/MS/MS (LCMS-8040; Shimadzu, Kyoto, Japan). Aqueous formic acid (adjusted to pH 3.0) mobile phase was delivered at 0.2 ml/min using a Develosil C30-UG separation column (5 μm, 2.0 mm × 250 mm; Nomura Chemical Co. Ltd, Tokyo, Japan). An equivolume amount of water was added to the plasma sample followed by twice the volume of methanol. After shaking vigorously, the mixture was centrifuged (21,880 ×g, × 5 min) to remove proteins. The supernatant (2 μl) was analyzed by LC/MS/MS. The optimized settings were as follows: ionization potential, −3.5 kV; collision potential, 11 V; heater temperature, 250°C.

MPO analysis with SDS-PAGE and Western blotting

To observe the up-regulation of MPO during pseudo-inflammation, SDS-PAGE and immunodetection analysis were carried out. The plasma samples were eluted with an equivolume of SDS-PAGE loading buffer containing 4% SDS, 20% glycerol, 100 mM Tris-HCl buffer (pH 6.8), and 12% 2-mercaptoethanol, and heated at 100°C for 3 min. Samples were separated by SDS-PAGE through a polyacrylamide gradient gel (5–20% acrylamide; ATTO, Tokyo, Japan). Proteins were transferred to a PVDF membrane and incubated with mouse anti-human MPO monoclonal antibodies (Cosmo Bio Co., Ltd., Tokyo, Japan) for 45 min at room temperature. Proteins were visualized with HRP-conjugated secondary antibodies (Bio-Rad Japan, Tokyo, Japan).

Results and Discussion

CAPD formation by reaction of CCPD and thiol compounds

The reaction of CCPD and GSH was conducted. GSH was added to a CCPD solution and MS spectra of the solution before (Fig. 2A) and immediately after (Fig. 2B) addition of GSH were measured by TOFMS with negative ionization. The deprotonated anion of CCPD (Found m/z −216.97421, Theoretical m/z −216.97701) observed before GSH addition disappeared immediately and an unknown ion was detected with an accurate m/z value of −183.01483 with a postulated molecular formula of C₅H₃N₄O₄ (Theoretical m/z −183.01543). Considering the structure of CCPD (C₅H₃N₄O₄Cl), the chemical structure of the unknown compound was thought to be CAPD (C₅H₄N₄O₄) in which the chlorine atom of the chloroamino group (-NHCl) in CCPD was replaced by a hydrogen atom. The CCPD and CAPD have a carbamic acid structure (=N-COOH) in their molecules. Carbamic acid is generally known to be unstable in aqueous solution to be decomposed to amine and carbon dioxide. Therefore, further investigation seems to be needed for the mechanism underlying the stability of CCPD and CAPD. Moreover, we think that their physical properties are also required to be examined in the future. Next, a Cys solution was induced into the CCPD solution at a constant rate, and the reaction mixture was analyzed by HPLC-UV. CCPD and CAPD were detected at 17 and 3.5 min retention times, respectively. CAPD formed within a brief period (20 min) after Cys induction, and after 80 min induction CCPD was completely decomposed (Fig. 2C). CAPD production correlated with CCPD degradation (Fig. 2D). The rates of CAPD formation and CCPD degradation were almost identical at approximately 0.25 μM/min, indicating that the reaction proceeded without formation of any stable intermediates. Considering that the rate of Cys induction was set to 0.5 μM/min, the stoichiometric number was calculated as 2.0, indicating that one molecule of CCPD reacted with two molecules of Cys to form CAPD and cystine. Indeed, the time course of cystine formation was similar to CAPD formation (Fig. 2D). These results were reproduced using GSH instead of Cys (Fig. 2E), suggesting that CCPD reacts with common thiol compounds to yield CAPD.

Fig. 2.

Formation of CAPD by the reaction of CCPD and thiol. MS spectra of CCPD (A) and CAPD (B) formed after addition of GSH to CCPD aqueous solution. Inserted numbers are accurate m/z values determined by calibration with TFA as an internal standard. (C) HPLC chromatograms monitoring the absorbance at 210 ‍nm of the reaction mixture 20 ‍min (upper) and 80 ‍min (lower) after continuous induction of GSH solution. Changes in concentrations of CCPD (), CAPD (), and cystine () during continuous induction of Cys (D) or GSH (E) to CCPD solution. Plots are expressed as mean ± SD (n = 3). (F) A plausible mechanism for conversion of CCPD to CAPD by the reaction with two thiol molecules.

Based on these results, a reaction mechanism was postulated (Fig. 2F). CCPD first reacts with a thiol (RSH), followed by HCl desorption to form an intermediate 1. However, the intermediate 1 is unstable and rapidly reacts with another RSH molecule to form CAPD and R-SS-R.

Stability of CAPD in phosphate buffer solution and human plasma

CAPD was dissolved in phosphate buffer solutions of various pHs (4.0–8.0) or collected human plasma and incubated at 37°C for 3 h. During incubation, the changes in CAPD concentration were measured by HPLC. The time courses of changes in CAPD concentrations in phosphate buffer solutions (Fig. 3A) and in collected human plasma (Fig. 3B) were plotted. The concentration of CAPD was relatively stable under neutral conditions and in human plasma for 3 h, suggesting that the formation of CAPD could be detected in vivo.

Fig. 3.

Stability of CAPD added to aqueous solutions and human plasma. (A) Changes in concentration of CAPD dissolved in phosphate buffer solutions of various pHs (4.0–8.0) during incubation at 37°C. (B) Change in CAPD concentration added to human plasma during incubation at 37°C.

UA oxidation induced by MPO system

MPO is released from activated neutrophils under inflammatory conditions and catalyzes the reaction of H₂O₂ and a halide ion to produce a hypochlorous ion. It is the predominant source of ClO⁻ in vivo. UA (400 μM) was oxidized with the MPO (250 mU/ml) system under co-existing NaCl (50 mM) and BSA (1.0 mg/ml). H₂O₂ was introduced at a constant rate (2.85 μM/min). CAPD was formed during incubation whereas CCPD formation was not observed (Fig. 4A). The CAPD formation occurred with UA degradation (Fig. 4B). Its yield, i.e., the percentage of CAPD formation to UA decrement, was 32% at 300 min. Considering that CCPD formation was estimated to be 30–50%,⁽³⁾ this value appears to be reasonable. This suggests that UA was oxidized by ClO⁻ generated from MPO to form CCPD, and the CCPD was subsequently converted into CAPD by reacting with thiol groups of BSA. Interestingly, PA and its precursor 4-hydroxyallantoin (4-HAL) were also formed during the oxidation. We previously demonstrated that these compounds are specific products of singlet oxygen (¹O₂).⁽²⁾ This indicates that ¹O₂ production occurred in the MPO-H₂O₂ system. In fact, there are many reports mentioning that the MPO-H₂O₂ system generates ¹O₂.^(4–7) A mechanism was postulated in which two-electron oxidation of H₂O₂ with ClO⁻ occurs to form ¹O₂. It has also been reported that heme-enzyme indoleamine monooxygenase (IDO) produces ¹O₂ under co-existence with H₂O₂ via the Fenton reaction and Russell mechanism.⁽⁸⁾ A similar heme-enzyme, catalase, is also reported to generate ¹O₂ during H₂O₂ elimination.^(9,10) MPO having a heme as an active center could produce ¹O₂ as do these heme-enzymes. In any case, there is no doubt that ¹O₂ was generated from the MPO-H₂O₂ system.

Fig. 4.

CAPD formation in oxidation of UA induced by MPO-H₂O₂ system. (A) Chromatograms of reaction mixture containing UA (400 ‍μM) and MPO (250 ‍mU/ml) at 0 ‍min (upper) and 60 ‍min (lower) during induction of H₂O₂ at a constant rate (2.85 ‍μM/min). (B) Time course of changes in concentrations of UA () and CAPD ().

CAPD formation during pseudo-inflammation

CAPD has been suggested to be the end-product of UA oxidation caused by ClO⁻ derived from MPO in the inflammatory condition in vivo. Hence, we induced pseudo-inflammation by LPS addition to freshly collected human blood. CAPD was clearly detected by LC/MS/MS at 2 h after LPS addition (Fig. 5A), whereas no CCPD was detected (data not shown). The CAPD concentration in the blood steadily increased over 6 h in association with MPO up-regulation in the presence of LPS (Fig. 5B). These results suggest that the following two reactions occurred: CCPD production caused by the reaction of UA with ClO⁻ generated from activated MPO, and subsequent conversion to CAPD by dechlorination of CCPD with thiol groups that were abundantly present in blood.

Fig. 5.

CAPD production during pseudo-inflammation induced by LPS addition into human blood. (A) LC/MS/MS chromatograms of human blood sample at 0 ‍h (upper) and 2 ‍h (lower) after LPS addition. (B) Changes in MPO expression in blood and time course of changes in CAPD concentration with () and without () addition of LPS.

Conclusion

In this study, we searched for a metabolite of CCPD that was eliminated upon addition to plasma, because CCPD was expected to be an indicator for ClO⁻ production in vivo. We found that CCPD was converted into CAPD by reaction with thiol groups. CAPD was relatively stable in human plasma and increased during MPO-induced UA oxidation and pseudo-inflammation in human blood initiated by LPS addition. Therefore, CAPD could be a biological marker for ClO⁻ generation in vivo.

Abbreviations

BSA

bovine serum albumin

CAPD

5-N-carboxyimino-6-aminopyrimidine-2,4(3H)-dione

CCPD

5-N-carboxyimino-6-N-chloroaminopyrimidine-2,4(3H)-dione

GSH

reduced glutathione

4-HAL

4-hydroxyallantoin

IDO

indoleamine monooxygenase

LPS

lipopolysaccharide

MPO

myeloperoxidase

PA

parabanic acid

TOFMS

time-of-flight mass spectrometry

UA

uric acid

Conflict of Interest

No potential conflicts of interest were disclosed.