A NOVEL NUCLEOTIDE FOUND IN HUMAN ERYTHROCYTES: 4-PYRIDONE-3-CARBOXAMIDE-1-β-D-RIBONUCLEOSIDE TRIPHOSPHATE Running title: Novel nucleotide in human erythrocytes

We report the identification of a hitherto unknown nucleotide that is present in micromolar concentrations in the erythrocytes of healthy subjects and accumulates at levels comparable with the ATP concentration in erythrocytes of patients with chronic renal failure. The unknown nucleotide was isolated and identified by liquid chromatography with UV and tandem mass detection, H nuclear magnetic resonance and infrared spectroscopy as 4-pyridone-3carboxamide-1-β-D-ribonucleoside triphosphate (4PYTP), a structure indicating association with metabolism of the oxidized nicotinamide compounds. Subsequently, we demonstrated formation of 4PYTP in intact human erythrocytes during incubation with the chemically synthesized nucleoside precursor (4-pyridone-3-carboxamide1-β-D-ribonucleoside, 4PYR). We noted preferential accumulation of monophosphate of 4PYR (4PYMP) over 4PYTP as well as a decrease in erythrocyte ATP concentration during incubation with 4PYR. Both the 4PYR phosphorylation and ATP depletion were blocked by an inhibitor of adenosine kinase. Plasma concentration of 4PYR was detectable but very low (0.013±0.006 μM) contrasting high daily urine excretion of this compound (26.7±18.2 μmol/24h) in healthy subjects, indicating much greater renal clearance than other nicotinamide metabolites, nucleosides or creatinine. We also noted 40-fold increase in 4PYR plasma concentration of patients with chronic renal failure (0.563±0.321 μM). We suggest that 4PYTP formation in the erythrocytes is a hitherto unknown process aimed to sequester potentially toxic 4PYR in a form that is subsequently released and excreted during passage of erythrocytes through the kidney. INTRODUCTION Nucleotides play a vital role in almost every process within living cells, including transfer of genetic information, energy metabolism and regulation. An optimal nucleotide pattern is essential for normal cell function and its disturbances leads to serious clinical consequences, covering the full spectrum of human disease . Changes in the nucleotide pattern are crucial to different phases of the cell cycle or differentiation, and pharmacological alterations of the nucleotide profile are components of anticancer, anti-inflammatory, immunosuppressive, antiviral and anti-ischemic therapies . The erythrocyte plays a central role in nucleotide metabolism related not only to their own requirements but also to the transport of nucleotide precursors and catabolites between their sites of formation, utilization and excretion . The chemical nature of all physiological nucleotides is believed to be well characterized . Therefore it was very surprising for us to note an unknown nucleoside triphosphate that was present at high levels in the erythrocytes of patients with renal failure, but also at lower concentration in the erythrocytes of healthy subjects . A correlation between concentrations of this unknown nucleotide in erythrocytes and the plasma concentration of a 1 http://www.jbc.org/cgi/doi/10.1074/jbc.M607514200 The latest version is at JBC Papers in Press. Published on August 17, 2006 as Manuscript M607514200 Copyright 2006 by The American Society for Biochemistry and Molecular Biology, Inc. by gest on N ovem er 7, 2017 hp://w w w .jb.org/ D ow nladed from nicotinamide catabolite 1-methyl-2-pyridone-5carboxamide (Met2PY) suggested that both compounds could be related . However, further comparison of chemical properties indicated that this is not the case. The present study was undertaken to fully characterize this novel nucleotide, to search for its possible precursors and to suggest its possible function. MATERIALS AND METHODS Collection of blood from patients and control subject and preparation of erythrocyte and plasma extracts To isolate and characterize the novel nucleotide, heparinized blood was obtained with written informed consent and Guy's Hospital Ethical Committee approval from adult patients with severe chronic renal failure attending renal out-patient clinics or hospitalized. All patients had end-stage renal failure (creatinine: 300-700 μmol/l) and were on either hemodialysis (HD) or on continuous ambulatory peritoneal dialysis (CAPD). The kidney function, defined as a glomerular filtration rate was less than 5 ml/min. For comparison, we used blood from adult controls (healthy laboratory staff). Plasma and erythrocyte extracts were prepared using trichloroacetic acid as we have described previously . Blood was centrifuged at 4000 g for 5 min immediately after collection. The plasma fraction was collected, and for erythrocyte extracts the top layer of packed cells comprising platelets and lymphocytes was discarded. The remaining erythrocytes were washed twice with 0.9% w/v saline, and 100 μl of packed washed erythrocytes was added to a 1.5 ml microcentrifuge tube containing 200 μl 10% trichloroacetic acid, and vortexed vigorously. After centrifuging for 2 min at 15,800 g, the supernatant was extracted with watersaturated diethyl ether. Extracts, if not used immediately were frozen and kept at -20oC. Urine collection for assessment of daily excretion was performed as we have described previously . Isolation and sequential degradation of the novel nucleotide The initial isolation of the novel nucleotide and its quantitative analysis were performed using the anion-exchange HPLC system described in detail previously . To obtain the nucleoside, nucleotide fractions collected by anion-exchange HPLC were pooled and concentrated by freeze drying. Incubation buffer (0.05 ml) containing 100 mM Tris/HCl, pH 8.0 with 50 mM MgCl2, was then added followed by 2 μl of enzyme solution containing 2 U of alkaline phosphatase. Incubation was carried out for 4h at 37oC. The reaction was terminated by addition of 0.2 ml of 10% trichloroacetic acid, and following centrifugation samples were extracted with diethyl ether as described for preparation of erythrocyte extracts. Post-reaction extracts were purified using the reversed-phase procedure described below, freeze-dried and subjected to chemical analysis by NMR or IR spectroscopy or further hydrolysed to obtain free base. The latter reaction was conducted by reconstitution of HPLC-purified and freeze-dried nucleoside in 60 % formic acid and incubation at 140oC for 1h . The post-reaction mixture was again freeze-dried, reconstituted in water and purified using the reversed-phase system. This method applied a Hypersil BDS 3 μm column (150mm/4.6 mm). Buffer A was 5 mmol/l ammonium formate; the mobile phase B was acetonitrile. A linear gradient from 100% A to 50% B in 12 min was used for elution, with 2 min of 50% mobile phase B running isocratically, and an additional 5 min for reequilibration with Buffer A before the next injection. The mobile phase was delivered at 1.2 ml/min. Known nucleosides and nucleobases were identified by their characteristic UV absorption spectra (range: 210-310 nm) and retention time compared with standards. The peak corresponding to the nucleoside and base peak in the post-reaction mixtures were identified in a chromatogram by its UV spectrum identical to nucleotide, by molecular weight and fragmentation pattern. The UV spectrum of the nucleoside was identical to that of the nucleotide, but the UV spectrum of the base peak was different. Anion exchange chromatography with mass detection The analytical system used was an LCQAdvantage or LCQ-Deca XP mass detector linked to a Surveyor chromatography system with an in line Agilent 1050 or 1100 diode array detector. Chromatographic procedure used chemically stable anion-exchange column (Phenomenex, 3 μm, LunaNH2, 150/2 mm). Buffer A was 5 mmol/l ammonium acetate; buffer B was 30 mM N'N'dimethylhexylamine/50 mM ammonium hydroxide delivered at a flow rate of 0.2 ml/min. A convex gradient profile from 100% buffer A to 100% buffer B in 12 min, was used for elution, with reequilibration time of 5 min with 100% buffer A. The unknown nucleotide peak was recognized by its characteristic UV spectrum. The mass detector was operated in negative ion mode. A Thermo-Finnigan Electrospray (ESI) ion source was used with 5 kV cone voltage setting and arbitrary nebulizing gas (nitrogen) flow set at 35%. The heated capillary was maintained at 220oC. Ion Optics parameters were optimized for ATP with standard instrument 2 by gest on N ovem er 7, 2017 hp://w w w .jb.org/ D ow nladed from routines. Helium was used as the collision gas. Collision energy of 28% was used for analysis of fragmentation pattern. Reversed phase chromatography with mass detection Analytical systems described above incorporating diode array and mass detection were used for characterization and quantitative analysis of the nucleoside and base in extracts and post-reaction mixtures. The reversed-phase method employed a Hypersil BDS 3 μm column, 150/2.0 mm. Buffers were running at a flow rate of 0.2 ml/min. Buffer A was 5 mmol/l ammonium formate; the mobile phase B was acetonitrile. A linear gradient from 100% buffer A to 50% B in 12 min was used for elution, followed by 2 min 50% B and 5 min re-equilibration at 100% A. Nucleosides and nucleobases were identified by their characteristic UV absorption spectra (range: 210-310 nm) and retention time compared with standards. The mass detector was operating in the positive ion mode. A ThermoFinnigan Electrospray ion source was used with 5kV cone voltage setting and arbitrary nebulizing gas (nitrogen) flow set at 35%. The heated capillary was maintained at 250oC. Ion Optics parameters were optimized for adenosine with standard instrument routines. Helium was used as collision gas. Collision energy of 35% was used to obtain the fragmentation pattern of the nucleoside under investigation. NMR and infrared spectroscopy A Varian Unity plus 500 MHz NMR spectrometer was used for nucleoside and base analysis with H NMR in D2O as solvent at room (22oC) temperature and with chemical shifts assigned according to the residual signal of water assuming its position at 4.64 ppm. The purified nucleoside was also analysed by infrared spectroscopy. Full details are provided in Supplementary Material. Chemical synthesis of pyridone derivatives Chemical synthesis of pyridone derivatives was performed as previously described: 4-pyridone3-carboxamide was prepared from commercially available 4-chloropyridine-3-carboxylic acid, and ribosylated under Vorbrüggen’s protocol . In the case of 4-pyridone-2-carboxamide , 2-picolinic acid was used as a starting material. Details of the chemical synthesis procedures, methods for structural characterizations and the spectral properties of intermediates and final products are provided in the Supplementary Material. Incubation of healthy human erythrocytes with 4PY and 4PYR Incubation of healthy human erythrocytes with 4PY and 4PYR was performed as we have described before . Erythrocytes of healthy humans were used for this experiment and were obtained and washed as described above. Erythrocytes were suspended in Hepes buffered Krebs medium at 20% haematocrit. 4PY or 4PYR were added at 30-1000 μM concentration as indicated in the legend to Figure 2. Incubation was carried out for 3, 6 or 12 h at 37oC. Incubation was terminated by the addition of trichloroacetic acid and extraction and analysis was completed as described in the first paragraph of the Methods section. Reagents Alkaline phosphatase was obtained from Amersham (U.K.). Chemicals for the synthesis were obtained from Lancaster and Aldrich. All commercially available nucleotides, nucleosides and bases were obtained from Sigma. Chromatographic columns were obtained from Phenomenex. HPLC grade solvents and buffer salts were obtained from VWR. Statistical analysis Data are presented as mean ±standard deviation (S.D.). Student’s t-test or one-way analysis of variance followed by Dunett test was used to compare two or more groups, respectively. P<0.05 was considered a significant difference. RESULTS AND DISCUSSION Identification of the unknown nucleotide Figure 1 presents the results of the key identification steps that revealed the structure of the novel nucleotide and allowed its identification as 4pyridone-3-carboxamide-1-β-D-ribonucleoside triphosphate (4PYTP). Liquid chromatography/mass spectrometry (LC/MS) analysis of the chromatographic peak corresponding to the novel nucleotide revealed a negative ion at m/z=509 corresponding to a molecular weight of 510. Similar analysis of ATP or GTP showed ions: m/z=506 and m/z=522 respectively, as expected. The fragmentation pattern obtained in MS2 and MS3 modes suggested that it is the base that is unique as only fragments larger than ribosetriphosphate were different in m/z ratio to fragments generated in MS2 and MS3 mode from ATP (not shown). This molecular weight was consistent with the suggestion that an oxidized metabolite of nicotinamide is the base constituent of the novel nucleotide, assuming that ribose is the sugar. To obtain the nucleoside constituent for analysis we treated the HPLC effluent fractions 3 by gest on N ovem er 7, 2017 hp://w w w .jb.org/ D ow nladed from containing the novel nucleotide pooled from several runs with alkaline phosphatase – an enzyme that nonspecifically releases nucleosides from nucleotides. The UV spectra of the nucleoside were identical to the parent nucleotide. LC/MS analysis of the nucleoside peak revealed a positive ion at m/z=271 corresponding to a molecular weight of 270, and analysis of its fragments in MS2 mode revealed ions at m/z=139 and m/z=122 that corresponded to sequential neutral loss of ribose and an amino group. Analysis with infrared spectroscopy and H NMR revealed further details of the structure. Our chemically synthesized 4-pyridone-3carboxamide-1-β-D-ribonucleoside (4PYR) was identical with the isolated biological compound with regard to chromatographic retention time, UV, mass, 1H NMR and infrared spectra. LC/MS analysis of the base peak obtained by acid hydrolysis of the nucleoside revealed a positive ion at m/z=139, corresponding to a molecular weight of 138. MS2 mode analysis revealed an ion at m/z=122 corresponding to neutral loss of an amino group. However, we had an insufficient amount of isolated material to perform heteronuclear 2D NMR analysis. To confirm its identity we therefore synthesized several of the most likely isomers suggested by 1H NMR analysis including 4pyridone-3-carboxamide (4PY), 4-pyridone-2carboxamide (4KP) and 2-pyridone-5-carboxamide (2PY). These chemically synthesized compounds were then compared with the base obtained from the novel nucleotide with regard to UV spectrum, chromatographic retention time and NMR spectrum. This indicated that synthesized 4PY is identical to the biologically isolated material. The base that we identified as a constituent of the novel nucleotide is biologically present only in the methylated form as 1-methyl-4-pyridone-3carboxamide (Met4PY) . The nucleoside 4-pyridone3-carboxamide-1-β-D-ribonucleoside (4PYR) has long been identified in human urine and plasma with most commonly used name ribosylpyridin-4-one-3carboxamide (PCNR) . Furthermore, the two anomers α-1 and β-1 of 4-pyridone-3-carboxamide ribonucleoside were described , but the UV and NMR spectra of α anomer were different from the nucleoside we isolated and synthesized. In addition isomers such as 2-pyridone-5-carboxamide-1-β-Dribonucleoside were also described . Changes in plasma 4PYR concentration or urine excretion received special attention in context of cancer treatment and were found to predict early death in patients active AIDS . Increase in plasma 4PYR could reflect cell damage, but on the other hand it may cause elevation of 4PYTP in cells of the immune system and contributes to loss of its function. Interestingly, nicotinamide is known to delay disease progression in AIDS . Since high-dose nicotinamide therapy emerges as an effective treatment in a variety of other pathological conditions such as diabetes, brain ischemia or renal failure it is important to take into account the effect of this treatment on cellular 4PYTP concentration as this may be related to therapeutic or toxic effects. Formation of 4PYTP in the erythrocytes and its effect on ATP concentration Identification of the structure and spectral properties allowed accurate quantitative analysis of the concentration of 4PYTP in erythrocytes in different clinical and experimental conditions. Concentrations of the 4PYTP, ATP and related metabolites in erythrocyte extracts of healthy subjects and patients with chronic renal failure are presented in Table 1. These data indicate that the healthy human erythrocyte concentration of 4PYTP is significant, corresponding to about 2% of the ATP level. Massive accumulation of 4PYTP was observed in patients with renal failure, up to 10% of ATP on average, or up to 30% in advanced chronic renal failure that is consistent with our previous suggestions . Figure 2 presents the results of the incubation of healthy human erythrocytes with 4PY or with 4PYR. Incubation with 4PYR resulted in progressive accumulation of 4PYTP while no change in 4PYTP was observed during incubation with 4PY or in controls (Figure 2A). Formation of 4PYTP from 4PYR in the erythrocytes was dependent on adenosine kinase activity as it was inhibited by its specific inhibitor 5’-iodotubercidin (Figure 2A). These results together with data from our earlier study provide evidence for the pathway by which 4PYTP is formed in the erythrocytes. Since specific inhibitor totally blocked this process, nucleoside phosphorylation mediated by adenosine kinase is the most likely step leading to incorporation of 4PYR into the nucleotide pool. The source of 4PYR in vivo is uncertain, but as mentioned above, this compound was identified in human plasma and urine before and our data shown in Table 2 confirm its presence in plasma in nanomolar concentrations in healthy subjects and demonstrate massive accumulation of 4PYR in patients with chronic renal failure. However, exact source and pathway leading to formation of 4PYR in plasma is less certain. We suggested earlier that the activity of aldehyde oxidase is essential for generation of the 4PYR since no 4PYTP accumulation was observed in an erythrocyte extract of a patient with molybdenum cofactor deficiency who developed chronic renal 4 by gest on N ovem er 7, 2017 hp://w w w .jb.org/ D ow nladed from failure . Since aldehyde oxidase is not present in the erythrocytes (liver is the major site of its expression in humans ) formation of erythrocyte 4PYTP may depend on the provision of 4PYR into the circulation by the other organs such as the liver. Most cells express full spectrum of enzymes for upstream and downstream metabolism of nicotinamide compounds and the exact nicotinamide-containing substrate which is oxidized remains to be identified. This question has been discussed broadly in our follow up-study in children with chronic renal failure (Synesiou et. al. submitted to Clin Sci) and considered by other authors . Significance of 4PYTP and its precursors Our discovery of 4PYTP in the human cells has several potential implications. Presence of this compound in the erythrocytes of healthy subjects at significant concentration indicates that this is part of a normal physiological mechanism. 4PYTP may be necessary for specific metabolic process. However, our data indicating that 4PYR phosphorylation into the monophosphate (4PYMP) occurs much faster than its subsequent metabolism into 4PYTP (Figure 2) seems to contradict this possibility. This rapid nucleoside phosphorylation and slow further metabolism within the erythrocytes suggests that this process is designed to remove 4PYR from the circulation, and not to make 4PYTP. 4PYR could be toxic for nucleated cells by interference ATP metabolism, by disruption of RNA or DNA synthesis either directly or after phosphorylation or through the potential for making an oxidized NAD by analogy with tyrazofurin and benzamide riboside. One evidence for potential toxicity is shown in our Figure 2B demonstrating decrease in erythrocyte ATP concentration during incubation with 4PYR. Although this occurred at relatively high concentration it may occur in vivo during prolonged exposures at lower levels or in specific cell types. Trapping of 4PYR in phosphorylated form within erythrocytes could prevent this potential deleterious process. Our results demonstrated massive accumulation of 4PYR in the plasma of patients with chronic renal failure (Table 2). This could be major factor that contributes to accumulation of the 4PYTP in the erythrocytes in patients with chronic renal failure. However, we made several additional important observations. The relative increase in plasma 4PYR in subjects with chronic renal failure by far exceeded that of Met4PY (Table 2) or any other known metabolite including creatinine. We noted more than 50-fold increase in plasma concentration of 4PYR comparing our group of patients with advanced renal failure to controls while increase in Met4PY was less than 10-fold, similar to relative increase in plasma creatinine in these patients (data not shown). Another important observation was disproportionally high excretion of 4PYR in urine compared to its plasma concentration in healthy subjects. Our measurements of both plasma concentration of 4PYR and its urinary excretion are close to earlier estimates of these values . Although plasma levels and urine excretion of 4PYR in our study were not performed in the same subjects to allow exact renal clearance calculations, even estimates indicate that this value is almost one order of magnitude greater than other nicotinamide metabolites or nucleosides. Renal clearance was found to be close to creatinine clearance in this study for Met4PY (data from Table 2), in our previous report for Met2PY or in studies of other authors for pseudouridine . Such a massive accumulation of 4PYTP in the erythrocytes and 4PYR in plasma of patients with chronic renal failure to levels that by far exceed what could be expected from reduced renal filtration suggests that there is an additional mechanism that enhances 4PYR excretion in healthy people. We could speculate that during passage of erythrocytes through the kidney there is a reverse process in which 4PYTP is broken down to 4PYMP and further to 4PYR. High local concentration of 4PYR would ensure its effective excretion (Figure 3.). Such a mechanism could explain the disproportionally high excretion of 4PYR in urine contrasting its extremely low plasma concentration in healthy adults. We could not exclude an active kidney excretion mechanism for 4PYR, but such a process has never been described for any nucleoside. An important practical aspect of such significant changes in plasma 4PYR in renal failure is that this measurement could become very sensitive and early marker of renal dysfunction. Although this hypothesis still needs further experimental evidence we propose that formation of the 4PYTP in the human erythrocytes is an element of a novel excretion pathway for oxidized nicotinamide metabolites. 5 by gest on N ovem er 7, 2017 hp://w w w .jb.org/ D ow nladed from