Product Studies and Mechanistic Analysis of the ... - ACS Publications

Jan 24, 2018 - After 4.5 h, yields of 2 increased, and no other products were observed (Figure 6). After 96 h, HPLC analysis showed 2 accounting for 8...
0 downloads 9 Views 2MB Size
Article pubs.acs.org/crt

Cite This: Chem. Res. Toxicol. 2018, 31, 105−115

Product Studies and Mechanistic Analysis of the Reaction of Methylglyoxal with Deoxyguanosine Sarah C. Shuck, Gerald E. Wuenschell, and John S. Termini* Department of Molecular Medicine, City of Hope and Beckman Research Institute, Duarte, California 91010, United States S Supporting Information *

ABSTRACT: Methylglyoxal (MG) is a highly reactive electrophile produced endogenously as a byproduct of glucose metabolism and protein catabolism and exogenously as a food contaminant. MG reacts spontaneously with proteins, lipids, and nucleic acids to form advanced glycation end products (AGEs), modifying or inhibiting their function. Protein AGEs are associated with pathological complications of diabetes, cancer, and neurodegenerative diseases, while the physiological impact of DNA, RNA, and lipid AGE formation is less well explored. Conflicting reports in the literature on the biologically significant DNA-AGE product distribution and mechanisms of formation prompted a re-examination of the reaction products of MG with dG, oligonucleotides, and plasmid DNA under varying conditions of MG:dG stoichiometry, pH, and reaction time. Major products identified using sequential mass fragmentation and authentic standards were N2-(1carboxyethyl)-2′-dG (CEdG), N2-(1-carboxyethyl)-7-1-hydroxy-2-oxopropyl-dG (MG-CEdG), and 1,N2-(1,2-dihydroxy-2methyl)ethano-2′-dG (cMG-dG). CEdG and MG-CEdG were observed in all DNA substrates, although cMG-dG was not detected to any significant extent in oligomeric or polymeric DNA. Product analyses of reactions under conditions of diminished water activity as well as results from H218O labeling indicated that MG hydration equilibria plays an important role in controlling product distribution. In contrast to previous reports, our data support independent mechanisms of formation of CEdG and cMGdG, with the latter kinetic product undergoing reversible formation under physiological conditions.



(pyruvaldehyde) at 65 °C for 18 h in sodium phosphate buffer (NaPi, pH 7) and reported the formation of a cyclic dihydroimidazolone 1, N2-(1, 2-dihydroxy-2-methyl)ethanodG, (cMG-dG, 2) as the only product.6 In contrast, others reported that reaction of 1 with a 20-fold excess of MG in NaPi (pH 7.4) at 37 °C resulted in formation of a novel product corresponding to the addition of 2 MG equivalents, N2-(1carboxyethyl)-7−1-hydroxy-2-oxopropyl-dG (MG-CEdG, 3), with no detectable formation of 2.7 Utilizing similar reaction conditions and analytical methods, Frischmann et al. observed adducts 3 and N2-(1-carboxyethyl)-2′-dG (CEdG, 4) in salmon sperm DNA incubated with excess MG, without any detectable formation of 2.5 Only when 1 was reacted with equimolar MG was 2 observed; however, 80% was found to decompose after 24 h,5 with 4 the sole remaining product after 7 days. Reaction of DNA with less than stoichiometric amounts of MG resulted mainly in the production of 4.5 Reaction of dG with a 3× excess of glyceraldehyde, which generates MG in situ following dehydration and enol-keto tautomerization, was reported to produce 4 in 10% yield after 2 weeks at 40 °C as the sole product.8 In contrast, the reaction of calf thymus DNA with

INTRODUCTION The Maillard reaction and the related chemistries of α-oxoaldehyde modification of proteins, lipids, and nucleic acids to yield advanced glycation end products (AGEs) have been studied since the early part of the 20th century; however there are still gaps in our knowledge regarding mechanisms of formation and their precise roles in human pathology. This is particularly true for the nucleic acid AGEs. Although several nucleic acid AGEs have been described, consisting predominantly of guanine adducts, there are still unresolved discrepancies in the literature regarding the relative stabilities of products, putative routes of formation, and product profiles under physiologically relevant conditions. Resolution of some of these issues will provide a clearer understanding of the potential role of nucleic acid AGEs in human pathology and could aid in the identification of potential biomarkers of metabolic disease. For example, DNA-AGEs have been shown to be elevated in diabetes, and their propensity to induce genomic instability suggests a potential molecular link between metabolic disease and enhanced cancer risk.1−4 Several groups have described products resulting from reactions of MG with dG and DNA, with conflicting observations.2,5−8 One of the earliest publications examined the reaction of dG (1) with an ∼30-fold excess of MG © 2018 American Chemical Society

Received: October 3, 2017 Published: January 24, 2018 105

DOI: 10.1021/acs.chemrestox.7b00274 Chem. Res. Toxicol. 2018, 31, 105−115

Article

Chemical Research in Toxicology

scavenge unreacted MG, followed by size exclusion chromatography using micro biospin columns (Bio-Rad). No artifactual nucleoside products potentially resulting from reaction with OPD were detected by either HPLC or MS. Double-strand (ds) DNA was prepared by mixing 1:1 stoichiometric equivalents of oligo A (sequence above) and oligo B, 5′-CTA ACG AGT CAG GCT-3′, and heating for 5 min at 90 °C followed by slow cooling to RT. DNA was then purified on a 12% nondenaturing polyacrylamide gel and isolated from gel slices by UV visualization and extraction with 45 mM Tris base, 45 mM borate, 1 mM EDTA, and 100 mM NaCl overnight at 4 °C. DNA was purified using a 0.2 μm syringe filter and adding 2× volumes of EtOH and ammonium acetate, pH 5.3, to a final concentration of 0.3 M. Samples were incubated at −80 °C for 1 h, after which they were spun at 13,000 ×g for 10 min. Supernatants were then removed, pellets dried, and samples resuspended in H2O. dsDNA was incubated with MG and processed as described above for ssDNA. DNA concentrations were determined by UV absorbance at 260 nm (Nanodrop, Thermo Scientific). pUC19 ds plasmid DNA (100 nM, kindly provided by Dr. Timothy O’Connor, City of Hope) was incubated with 0.1 mM MG (∼1 × 103 fold excess) as described above for ssDNA. M13mp18 ss plasmid DNA (100 nM) was similarly reacted with 0.1 mM MG. DNA samples were diluted to 0.5 μg/mL and heated at 95 °C for 5 min followed by incubation on ice. Following sample cooling, ammonium acetate (pH 5.3) was added to a final concentration of 10 mM along with two units of Nuclease P1 (US Biological). Samples were incubated at 45 °C for 2 h followed by addition of ammonium bicarbonate (pH 8.0) to a final concentration of 100 mM and 0.002 units of Phosphodiesterase I (Sigma), with continued incubation at 37 °C for 2 h. Nucleotides were dephosphorylated with calf intestinal phosphatase (0.5 units, New England Biolabs) overnight. HPLC Analysis of Nucleosides. Samples were analyzed using an Agilent 1100 HPLC system (Agilent Technologies) equipped with a 10 × 250 mm, 5 μm XBridge Prep C18 column (Waters). Products were monitored by UV absorbance at 260 nm and separated using mobile phase A (H2O with 0.1% formic acid) and mobile phase B (acetonitrile with 0.1% formic acid). The following gradient was employed: 0 to 9% B, 15 min; 9 to 9.5% B, 40 min; 9.5 to 90% B, 5 min; hold at 90% B for 10 min, return to 0% B in 5 min, and hold for 5 min. Chromatograms and peak area integrations were analyzed using Agilent Chemstation software. Mass Spectrometry. Products isolated by HPLC were directly infused into a Thermo Scientific hybrid linear ion trap-Fourier transform (LTQ-FT) ion cyclotron resonance mass spectrometer in positive ion mode. Fragmentation of precursor ions was performed by collision-induced dissociation (CID) with helium gas using energies ranging from 10 to 50 eV. Xcalibur software (Thermo Scientific) was used for data analysis. For selected reaction monitoring (SRM) and neutral loss analyses, an Agilent 1290 UHPLC system was used in conjunction with an Agilent 6490 QQQ mass spectrometer. Analyte separation was performed using an Agilent SB-Aq column (2.1 × 50 mm, 1.8 μm) using the following solvents: mobile phase A, 0.1% formic acid in H2O; mobile phase B, 0.1% formic acid in acetonitrile. The gradient program ramped from 3% B to 10% B over 3 min; 90% B in 1 min, with a return to 3% B over 1 min at a flow rate of 0.4 mL/ min. SRM was used to detect mass transitions in positive ion mode m/ z 340 to 224 (R and S-4; 2) and m/z 412 to 296 (3). High-resolution MSn analyses were performed on a Thermo Orbitrap mass spectrometer utilizing a heated electrospray ionization source with analytes detected in positive ion mode. Analytes were separated with an Agilent SB-Aq column (2.1 × 50 mm, 1.8 μm), with identical mobile phases and gradient as described above with a Thermo EasyNano 1000 LC system. DNA adducts were identified by monitoring the neutral loss of 2′-deoxyribose (116 amu). Isotopic standards were synthesized as previously described.11 MG Reaction with CEdG. To analyze products formed upon reaction of MG with CEdG, 1 mM R-CEdG (synthesized as previously described)11 was incubated with 1 mM MG in NaPi, pH 9.5 for 24 h, followed by product analysis using HPLC and LTQ mass spectrometric analysis as described above.

MG in buffer containing Tris, desferrioxamine (DFOM), and EDTA was reported to yield predominantly cMG-dG (2), with CEdG (4) identified as a minor product.2 In these studies, 2 was reported to have a half-life of